Utah State University DigitalCommons@USU
All Graduate Theses and Dissertations Graduate Studies
5-1966
Cholesterol Levels in Serum and 5β – Pregnane – 3α,20α - diol in Urine of University Students
Gertrude Kuei-Shu Chiang Utah State University
Follow this and additional works at: https://digitalcommons.usu.edu/etd
Part of the Food Science Commons
Recommended Citation Chiang, Gertrude Kuei-Shu, "Cholesterol Levels in Serum and 5β – Pregnane – 3α,20α - diol in Urine of University Students" (1966). All Graduate Theses and Dissertations. 4820. https://digitalcommons.usu.edu/etd/4820
This Thesis is brought to you for free and open access by the Graduate Studies at DigitalCommons@USU. It has been accepted for inclusion in All Graduate Theses and Dissertations by an authorized administrator of DigitalCommons@USU. For more information, please contact [email protected]. CHOLESTEROL LEVELS IN SERUM AND 5� -PREGNANE-3D( ,20o(.-DIOL
IN URINE OF UNIVERSITY STUDENTS
by
Gertrude Kuei-Shu Chiang
A thesis submitted in partial fulfillment of the requirements for the degree
of
MASTER OF SCIENCE
in
Nutrition and Biochemistry
UTAH STATE UNIVERSITY� Logan, Utah
1966 TABLE OF CONTENTS
INTRODUCTION 1
REVIEW OF LITERATURE 4
Physiological effects of progesterone 4
Source and metabolism 4
Function in the body and its excretion 6
Analytical method of urinary pregnanediol
The relationships of serum cholesterol and coronary atherosclerosis 8
The influence of the sex hormones on the circulating lipids . 13
Serum cholesterol levels and eating frequency 14
Analytical method of serum cholesterol 16
METHOD AND PROCEDURE 18
Experimental design 18
Collection and storage of urine specimens 20
Collection 20
Storage . 22
Analysis of ur inary Sft-pregnane -~, 2~-diol 22
Method 22
Equipment 22
Reagents 23
Procedure 23
Hydrolysis 24
Extraction 24 llage
Washing and drying 24
Preparing the glass plates 25
Chromatography 26
Spectrophotometry 28
Calculation 29
Recovery 29
Preparation of serum samples 29
Analysis of serum cholesterol 31
Method 32
Equipment 32
Reagents 33
Procedure 34
Extraction and sampling 34
Precipitation of total cholesterol 34
Precipitation of free cholesterol . 35
Washing of the cholesterol digitonide precipitates 35
Color development 36
Calculation 37
Recovery 37
RESULTS AND DISCUSSION 39
Urinary pregnanediol levels 39
Effect of menstrual cycle 43
Effect of two meals versus three meals per day 43
Serum cholesterol levels 55
Effect of menstrual cycle 63 Page
Effect of two meals versus three meals per day 63
Effect of dietary factors 66
Relationships between serum choles terol and degra dation products of androgens and progesterone 72
COMMENTS AND CONCLUSIONS 76
SUMMARY 78
LITERATURE CITED 80
APPENDIX 84 LIST OF TABLES
Table Page
1. Experimental design 19
2. Mean age, weight, and height of subjects 21
3. Recovery test of urinary pregnanediol 30
4. Recovery test of serum cholesterol 38
5. Summary of mean squares obtained by analysis of variance of data on urinary pregnanediol 40
6. Mean values of pregnanediol in urine of subjects consuming two meals and three meals per day 41
7. Urinary pregnanediol values of normal humans as reported in the literature . 42
8. Summary of mean squares obtained by analysis of variance of data on serum cholesterol of male and female subjects 56
9. Summary of mean squares obtained by analysis of variance of data on serum cholesterol of female subjects 57
10. Percentage distribution of nine subjects within various ranges of serum total cholesterol values 58
11. Mean serum cholesterol concentrat ion by sex and meals
12. Mean values of total and free cholesterol in serum of subjects consuming two and three meals per day 61
13. The range of serum total cholesterol values for individual subjects by number of meals . 62
14. The effect of menstrual cycle on the concentration of the serum cholesterol
15. Mean food intake of the subjects 70
16. The relationship between serum cholesterol and urinary neutral 17-ketosteroid, and pregnanediol by sex 75 Table Page
17. Urinary 5$ -pregnane3oL,20ol-diol values for individual subjects consuming two meals and three meals per day
18. Urinary pregnanediol excretion values for the individual subjects consuming two meals and three meals per day by day of menstrual cycle
19. Urinary pregnanediol excretion values for the individual subjects by day of menstrual cycle 87
20. Serum total cholesterol values for individual subjects eating two meals and three meals per day 88
21. Serum free cholesterol values for individual subjects consuming two meals and three meals per day LIST OF FIGURES
Figure Page
1. Biosynthesis and metabolism of progesterone
2. The effect of day of menstrual cycle on excretion values of pregnanediol and the levels of serum cholesterol 44
3. The effect of day of menstrual cycle on mean excretion values of pregnanediol of 5 subjects 44A
4. Urinary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycle for subject S.R. 45
5. Urinary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycle for subject T.D. 46
6. Urinary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycle for subject J .A. 47
7. Urinary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycle for subject B.N. 48
8. Urinary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycle for subject M.J. 49
9. The effect of number of meals on mean excretion values of urinary pregnanediol for male subjects 50
10. The effect of number of meals by day of test period on pregnanediol excretion values for male subjects 51
11. The effect of number of meals on mean excretion values of urinary pregnanediol for female subjects 53
12. The effect of number of meals by day of menstrual cycle on pregnanediol excretion values 54
13. Mean total cholesterol values for men with or with- out lunch by days of study . 65 Figure Page
14. Mean total cholesterol values for women with or with out lunch by days of menstrual cycle 67
15. The effect of number of meals by days of study on total serum cholesterol values fo r men 68
16. The effect of number of meals by days of menstrual cycle on total serum cholesterol values 69
17. Mean serum cholesterol and urinary neutral 17- ketosteroid and pregnanediol values by days of study for men 73
18. Mean serum cholesterol and urinary neutral 17- ketos teroid and pregnanediol values by days of menstrual cycle for women ACKNOWLEDGEMENTS
The author wishes to express her gratitude to Dr. Ethelwyn B.
Wilcox for her untiring efforts in initiating this problem and in
making suggestions and counseling throughout this study. Appreciation
is also expressed to Dr. Joseph C. Street, Mrs. Grace J. Smith, and
Dr. Elveda Smith for their special assistance and suggestions. Thanks
are given to Dr. Donald V. Sisson for his help with the statistical
analysis.
This study was part of the Western Regional Research Project on
Biological Interrelationships in Lipid Metabolism of Importance to
Man. It was supported in part by W-44 funds obtained under the
Research and Marketing Act.
Acknowledgement is given to all others who gave their assistance
and encouragement and thanks to those persons who acted as experimental subjects. INTRODUCTION
Since there is a high incidence of coronary heart disease in
humans of middle-age and older in the United States, and the mortality
of this disease is still increasing, a great deal of research has been done in this field. Although many theories have been formulated con cerning the cause of atherosclerosis, it is a very complex disease, and appears to be influenced by many factors. Disturbed lipid meta bolism is widely believed to be involved in the development of the vascular lesions found in arteriosclerosis, although the nature and origin of the disturbance is unknown. The mechanism for the elevation of serum cholesterol and triglyceride levels is not known; but sex hormones and dietary constituents are believed to be among the factors involved.
The concept that atherosclerosis is a metabolic disorder involving lipids and lipoproteins has stimulated extensive investigation of the endocrine influences on circulating lipids and on lipid metabolism.
The results have consistently indicated that estrogens decrease cir culating cholesterol and prevent cholesterol induced coronary atheros clerosis, while androgen administration tends to increase circulating cholesterol . There are relative few studies of the effect of other sex hormones, such as progesterone. There is only limited information on the relationship between the level of progesterone and the concentra tion of serum cholesterol in healthy adults while living under ordinary living conditions. Further work with more subjects is desirable. Since the discovery by Marrian and co-workers (1929) of S~preg
nane-3ol,200l-diol in urine of pregnant woman and the perfection of a
quantitative method for its determination by Venning (1937), this compound
has received wide attention because of its close relation to the meta-
holism of the corpus luteum hormone, progesterone. Pregnanediol is the
major end product of progesterone catabolism and its estimation provides
a useful index of luteal function. Progesterone has been isolated
from three mammalian tissue sources, the corpus luteum, the adrenal, and
the placenta. The function of progesterone i s to promote the prolifer
ation of uterine mucosa and thus to prepare this tissue to receive the
fertilized ovum.
Cholesterol, literally meaning bile solid-alcohol, derives its
name from the fact that it was first isolated from human gallstones,
of which it is generally the chief component. The amount of chole
sterol in animal tissues varies widely. It is particularly abundant
in brain and nerve tissue, adrenal glands , and egg yolk.
It had been shown that the administration of deuterium-labeled cholesterol to a pregnant woman gave rise t o labeled pregnanediol in
the urine. Presumably, the administered cholesterol was converted in
the placenta to progesterone, from which pregnanediol was then formed
(Fruton et al., 1958). Subsequent work demonstrated that adrenal 14 tissue can convert c -labeled cholesterol to labeled progesterone, as well as to corticosterone and cortisol, and that cholesterol is a more efficient precursor of the hormones than is acetate.
Observations in animals limited to meal eating rather than ad libitum feeding, have shown that serum cholesterol was significantly raised (Cohn et al., 1962). The rats trained to eat their food in a 3
short period each day also showed ma rkedly increased lipid synthesis .
The enhanced lipogenesis resulted in an increase in fat ceposition
in adipose tissue. These findings have been interpreted to suggest
that using frequent small feedings might prove to be beneficial to
people who have abnormal lipid metabolism.
This study is part of a larger problem whose purpose was to
determine relationships that exist between serum cholesterol and
concentrations of estrogens and the degradation products of andro-
gens, adrenal steroid hormones, and progesterone in urine of healthy
young adults consuming self-selected diets under home living conditions.
Simultaneous studies of these factors in human subjects have been very
limited. The objectives of this study was t o determine the levels of serum cholesterol and urinary excretion values of pregnanediol, and any relationships between the two biochemical indices that might exist.
The research was based on a group of university students (five women and four men) maintained on self-chosen diets who were eating two meals per day with no lunch or three meals a day. Urine spec imens were collected for hormone estimation and finger-tip blood samples for cholesterol determination. Dietary records were also collected for dietary calculation.
Chemical analyses of free and total cholesterol were made by using the method of Galloway et al. (1957). Determination of pre nanediol in urine was made by modification of the method of
Eberlein and Bongiovanni (1958) on thin-layer chromatography. REVIEW OF LITERATURE
Physiological effects of progesterone
Source and metabolism
The progestational hormone, progesterone, may be regarded as a
derivative of "pregnane." It may be sedignated "4-pregnene-3,20-
dione." It has methyl groups at C-10 and C-13, and containes 21 C
atoms. Figure 1 shows the biosynthesis and metabolism of progesterone.
The chief source of progesterone i n ·~ the body is the corpus luteum, but it can be demonstrated in the placenta and the adrenals also.
Progesterone may originate from cholesterol, although this is not necessarily an immediate precursor. Acetate may be the only obligatory precursor (Cantarow et al., 1962). Comparatively little is known regarding its intermediary metabolism, although there is evidence that the liver plays an important role in this connection, as in the case of estrogens and androgens. Progesterone metabolites have been found in the bile; the liver appears to be involved in the removal of this hormone from the circulation, in its degradation to unidentified compounds, and in its reduction to pregnanediol, which is biologically inactive. The latter is then conjugated with glucuronic acid, passed into the systemic circulation, and excreted in the unine.
The first significant experiment in this field was the adminis tration of deuterium-labeled cholesterol to pregnant women with the demonstration of deutero-pregnane-3ot,20Dl- diol in the urine by Bloch
(1945). This conversion of cholesterol to pregnane-3ot,20C(-diol Cholesterol isocaproic acid Pregnenolone (5-pregnene-3-ol-20-one)
Testosterone
Adrenocortical hormones Progesterone !
Pregnane
CCONa Pregnanolone I (pregnane-3-ol-20-one) HC~ CH3 H~OH 0 I I H?C;J HCOH CH3 m Glucuronide I c 0 -CCJ:t-----/::..__ Pregnanediol glucuronide Pregnanediol
Figure l. Biosynthesis and metabolism of progesterone demonstrated by indirect means that cholesterol can be converted to steroid hormones . The finding of deutero-pregnane-3ol ,20ol-diol in urine was presumptive evidence that progesterone had been formed from the cholesterol and that the progesterone so formed was metabolized to the diol. Since these early studies of Bloch a variety of experi ments have appeared in the literature during the past 10 years. Fruton et al. (1958) found the same result and indicated that adrenal tissue can convert c14-labeled cholesterol to labeled progesterone, as well as to corticosteroids, and that cholesterol is a more efficient pre cursor of the hormone than is acetate.
Function in the body and its excretion
Progesterone is secreted by the corpus luteum of the ovary during the period of its functional activity. It is concerned in the latter half of the menstrual cycle, mainly with preparing the endometrium for nidation of the fertilized ovum if concepation has occurred.
During early pregnancy, it is produced by the corpus luteum of preg nancy, and perhaps by other ovarian luterinized cells, and later by the cyncytial cells of the placenta. During gestation, progesterone in association with estrogen, appears to be concerned with maintenance of a state of quiescence of the uterine muscle. Progesterone also causes an increase in glycogen, mucin, and fat on the uterine lining epithelial cells. The sudden decrease in progesterone and estrogen leads to a series of changes culminating in bleeding or menstruation.
It modifies the action of estrogen on the vaginal epithelium during the menstrual cycle. In conjunction with estrogen, progesterone causes development of the alveolar system of the breasts and sensiti zes them for the action of lactogenic hormone. It counteracts the effect of estrogen on the fallopian tubes and uterine cervix. In large doses
it exerts androgenic effects, perhaps by conversion to androgenic
metabolites. Progesterone is responsible for the rise in basal tem
perature which occurs during the corpus luteum phase of the normal
menstrual cycle. This is believed to be related to an increase in
the basal metabolic rate.
Human urine contains only biologically inactive progesterone metabolites, chiefly the glucuronidate of pregnanediol. Other isomers
present especially during pregnancy, but in smaller amounts, are allo
pregnane-3()(. ,20ol-diol and allopregnane-3,(3 ,200(-diol. Pregnanediol glucuronidate occured in the urine during the luteral phase of the menstrual cycle and attained its maximum value between days 20 to 25 of the menstrual cycle (Zarrow et al., 1964). Much less is excreted during the follicular phase. The amount of pregnanediol present has been considered as a measure of corpus-luteum activity. However, marked variations occur, due to the variable ovarian endometrial, hepatic, and renal factors which influence it excretion.
Analytical Method of Urinary Pregnanediol
Many methods have been devised to measure pregnanediol in urine, both as sodium glucuronidate obtained by enzyme hydrolysis of the conjugate (Venning, 1937; Allen and Viergiver, 1941) and as the free alcohol, obtained by acid hydrolysis of the conjugate (Astwood and
Jones, 1941). In an attempt to improve the specificity of earlier methods, de Watteville (1951) and Stimmel et al. (1952) introduced alumina column chromatography of the crude urinary extracts. Klapper et al. (1955) also reported a more elaborate procedure, which employed 8
dolumn chromatography before and after acetylation of extracts. A
method which uses paper chromatography described by Eberlein and
Bongiovanni (1958) is probably the most specific and most sensitive
method for estimation of pregnanediol in small volumes of urine.
Recently Waldi and Munter (1960), and Bang (1964) reported a rapid and
simple method by using thin-layer chromatography to determine urinary
pregnanediol.
A modification of the method of Eberlein and Bongiovanni (1958) in
which paper chromatography was changed to thin-layer chromatography was successful and was used in this study. The merits of thin-layer
chromatography have been stated by Mangold (1961) and Randerath (1963)
to be rapidity and reproducibility.
The Relationships of Serum Cholesterol and Coronary Atherosclerosis
Atherosclerosis is a complex disease of the arteries. It is known
that a number of factors influence or are related to its development, although the exact cause which initiaties the formation of the atheroma is not clear . These factors include the following: a high content of cholesterol in the blook, elevation of blood pressure above normal, kidney damage, presence of diabetes, heredity, hypothyroidism, obesity, and a habit of heavy cigarette smoking (Stamler et al., 1963). Age, and sex may· also be involved.
For a long time it has been known that cholesterol and other lipids are found in the plaques in atherosclerosis and coronary heart disease. Some workers have found that serum cholesterol levels are significantly higher in coronary patients than in controls (Gertler and Garn, 1950; Lawry et al., 1957: keys and Fidanza, 1960). 9
Albrink (1965) also indicated that it is unquestionably that cholesterol concentrati on i s associated with coronary artery disease, however, the overlap is great and the difference bet ween the normal and coronary population almost disappears after age fifty when the peak incidence of coronary artery disease occurs .
Several methods designed to reduce the amount of cholesterol i n the blood have been suggested (report by the American Heart Associ~ ations, 1961).
First, it could seem that the simplest way t o reduce cholesterol in the blood is to eat less foods containing cholesterol , but the problem is much more complex than this. If the amount of cholesterol in the diet is ma rkedly decreased, but the calorie intake kept constant, the body will make mo re cholesterol from other substances, chiefly from other types of fat. Sometimes this is nearly enough to make up for that which has been removed from the diet (American Heart Associations,
1961).
Second, reduction of the total calorie intake by decreasing the amount of ordinary fat in the diet, usually causes reduction of the block cholesterol concentration. Avoidance of excess fat in the diet also helps avoid obesity, because one gram of fat provides 9 calories, while one gram of protein or carbohydrate provides only 4 calories each. This does not mean that unlimited amounts of carbohydrate and protein should be eaten, for these, in excess, may be converted into cholesterol and which also may lead to obesity which in turn increases the level of cholesterol in the blood.
Third, the blood cholesterol concentration may also be reduced by controlling the amount and type of fat in the diet without altering 10 calor ic intake. No t all fats i n the diet have the same effect on the amount of cholester ol i n the blood. Those high in saturated fat tend to increase the choles t e rol in the blood, those high in mono-unsaturated f atty acids have little effect, and those high in poly-unsaturated fatty acids decr ease cholesterol.
Anot her appr oach to r educe the cholesterol level in blood is from the vi ew point of the solubility of cholesterol in the fat as affected by the intestinal absorption and reabsorption of cholesterol. Cholesterol is absorbed only in a solution in fat which is itself in the process of abs orpti on. Endogenous cholesterol is similarly affected (Wilkens et al.,
1963). Th e amount of dietary fat or the solubility of cholesterol in the dietary fat may thus become the limiting factors in cholesterol absorption and reabsorption. Wilkens et al. (1963) thus predicted:
1. tha t the lowest serum cholesterol will occur when a fat-free
d i et i s fed .
2. that above a certain critical quantity of fat for solubiliza
tion of the endogenous and exogenous cholesterol, no further
increase in serum cholesterol will be found.
3. that a plateau is serum cholesterol level will be reached as
increasing quantities of exogenous cholesterol are added to
a given dietary fat .
4. that decreased absorption of cholesterol resulting from feeding
a fat with a low cholesterol dissolving power will cause in
creased liver synthesis of cholesterol and consequently greater
excretion of cholesterol as well as its metabolites, including
bile a cids.
The mechanism is probably oversimplified and the solubility tests 11 highly unphysislogical . The intestinal lipids subject to absorption are believed to exist largely in the form of mixed micelles composed of bile salts, monoglycerides, fatty acids, and nonpolar lipids. Such micelles in the upper intestine might compete with emulsified lipids
(larger, unabsorbed particles) for a nonpolar lipid such as cholesterol and therefore affect its absorption. Unquestionably, the solubility of cholesterol in the intestinal lipids is much more complicated than simple solution of cholesterol in the dietary fat (Wilkens, 1963).
The concentration of cholesterol in serum is the result of the amount ingested and absorbed plus the amount synthesized, as related to the amount degraded to bile acids and excreted in the bile as such, or as cholesterol, and the amount utilized or stored in the tissues.
Some of the bile acids and cholesterol are reabsorbed from the intes tine and act as elements of a feed-back mechanism, influencing the amounts of cholesterol synthesized and degraded. The exact details of this mechanism have not been elucidated (Goldsmith, 1964).
In the popular mind, cholesterol and atherosclerosis are regarded almost as synonymous. A connection between the two seems definite, but the extent of the relationship, the question of endogenous versus exogenous cholesterol, the influence of other factors, and the pathology of the disease are all germane to the problem. The mechanism for the association of clinical coronary disease with a disturbance of the plasma lipids might be as follows (Oliver and Boyd, 1958):
Research workers do not believe that cholesterol initiates changes in the artery wall but after an injury does occur within the wall of the artery, cholesterol and other lipids are deposited at the point of injury and a disturbance of the plasma lipids occurs. There is much experimental 12 evidence that in mammals the plasma lipids originate almost exclusively from hepatic synthesis. These lipids are discharged into the circulation and remain there for a finite period before being degraded in the reticuloendothelial system, the liver, or elsewhere. Thus, it would be necessary to argue that any primary intimal lesion in the coronary arteries has a "secretory" role and is capable of influencing one or more of these processes.
Some degree of hypercholesterolemia seems to occur in the produc tion of atheroma in animals. The extent and duration of the hyper cholesterolemia appears to determine the severity of the atherosclerotic lesions. There is a great deal of evidence that, under physiological conditions, plasma cholesterol passes through the vessel wall from the lumen. Disturbance of this process may lead to the accumulation of cholesterol in the intimal atherosclerotic plaque. Arterial lesions have been produced by the parenteral administration of cholesterol emulsions, and the intravenous administration of hypercholesterolemic plasma from cholesterol-fed rabbits to normal rabbits (Oliver and Boyd,
1958).
Albrink (1965) postulated that it is quite possible that serum cholesterol and serum triglycerides together operate to accelerate the atherosclerotic process and that the effects of one cannot be separated from the effects of the other. On the other hand, the value of cholesterol in predicting coronary artery disease could conceivably be due to its value in predicting triglyceride elevation. The Influence of The Sex Hormones on the Circulating Lipids
Evidence is accumulating that the gonads exert an important in fluence on the level of circulating lipids and lipoproteins (Adlersberg,
1958). The comparatively rare occurrence of coronary artery disease in women during the reproductive phase is well known. Bilaterally ovari ectomized women show a higher incidence of coronary artery disease than normal women of corresponding ages.
Depression of the plasma cholesterol, plasma cholesterol:phospho lipid ratio, and theJS-lipoprotein cholesterol with elevation of the ol-lipoprotein cholesterol have been found to occur regularly at ovula tion. Changes in lipid concentrations in the opposite direction occured during the luteal phase of the cycle. The secretion of estrogens also undergoes regular cyclical variation and is generally believed to rise to a maximum at or about the time of ovulation, with a possible peak just before menstruation. It has been suggested that the depression of certain plasma lipids at ovulation could result from the increased endogenous secretion of estrogens at that time and that the opposite changes observed during the luteal phase, despite the second rise in estrogen excretion, might possibly be related to anta gonism by other hormones (Aldersburg, 1958).
The administration of progesterone does not cause any marked elevation of the circulating lipids and lipoproteins (Oliver and Boyd,
1956a). Androgens have been found to elevate serum lipids (Barr, 1953;
Oliver and Boyd, 1956a; Aftergood et al., 196~. Serum Cholesterol Level and Eating Freguency
A change in body metabolism associated with a difference in the
frequency with which the day's allotment of food was ingested was first
noted by Tepperman et al. (1942). These workers trained the laboratory
rat, an animal that usually eats small quantities of food on many occa
sions over twenty four hours, to consume its ration in a daily 1 to 2
hour period. The new eating habit resulted in an increase of the RQ
sufficiently great to be interpreted as an indication of increase in
lipogenesis.
Some of the evidence suggesting feeding frequency to be a signi
ficant factor in the regulation of intermediary metabolism is based on
the fact that the activities of a number of enzymatic pathways have
been shown to be influenced by the load of nutrients (substrates) pre
sented to them per unit of time . It follows, therefore, that the
alterations in body metabolism seen with different eating patterns
could conceivably result from the limited activities of some enzymatic pathways with consequent adaptation of alternate ones (Cohn, et al.,
1964).
The animal experiments appear to have yielded clear- cut results in demonstrating that feeding frequency plays a significant role in
the regulation of serum cholesterol levels and in the induction and regression of experimental atherosclerosis (Okey et al., 1960 ;
Hollifield, 1962; Rakes, Lister, and Reid, 1961; Cohn et al., 1961).
Chickens and rabbits restricted to eating a diet containing modest amounts of cholesterol, for 2 hours per day, exhibited double the serum cholesterol levels and four to seven times the incidence of atherosclerotic lesions than was shown by animals consuming 30 per cent
more diet over a 24-hour period. The monkey and female rat likewise
respond to a restriction of the time during which food is available by
an increase in the levels of serum cholesterol.
Man's reactions to different eating habits, in regard to serum
lipid levels, appear to be the same as that of other species. One
subject with idiopathic hyperlipemia and hypercholesterolemia reacted
to a change from a meal eating to a nibbling regimen with a drastic
fall in his elevated serum lipid levels (Cohn, 1961) . Hashim et al .
(1960) have noted that subjects placed on a 6-times-a-day formula diet,
after having been on a 3 meals a day hospital diet, responded with
decreased serum cholesterol levels, regardless of the fat in the diet.
The type of fat in the diet, however, did influence the magnitude of
the fall.
Ende's (1962) volunteer subjects also showed a general tendency
for the total serum cholesterol to rise during starvation, particularly
on the third and fourth day. Yo ung men below the age of 35 had the
highest mean rise in serum cholesterol. The least elevation occurred
in older men, particularly those with known atherosclerosis. In normal
women below the age of 35 the average rise in serum cholesterol was
slightly less than in men of a similar age gr oup.
The mechanisms that could account for the elevated serum cholesterol
levels, as influenced by the periodicity of food intake, are numerous but no experiments have been reported to elucidate any of the possibi lities. Cohn (1964) has speculated that feeding frequency could in fluence the absorption of cholesterol f r om t he gastr ointestinal t r act ,
the enterohepatic circulation of cholesterol, the degradat ion of 16
cholesterol bo bile acids or change the gastroint estinal f lora with a
resultant effect on these processes .
The data from a review of the literature indicate that the manner
in which man responds to different f eeding frequencies have not yielded
as well defined results as those from the experimental animals. The
results now available are s uggestive, however, in indicating that man
reacts like other species . More data on this aspect of cholesterol
metabolism a nd controlled feeding are obviously needed.
Analytical Method of Serum Cholesterol
The most widely used of all methods of cholesterol identif ication
is the reaction initially described by Liebermann in 1885 and developed
by Burchard; it bears both their names. The classical Liebermann
Burchard reaction entails treatment of a solution of cholesterol in
acetic anhydride with concentrated sulfuric acid. There is an ensuing
color display which goes from red to violet to green (Kritchevsky,
1956).
In 1934, Schoenheimer and Sperry published the s o called "micro" method for the determination of cholesterol which has become the s tandard procedure in most laboratories. This "micro" method used
0.2 ml of serum for the free and combined cholesterol analyses . In
this method cholesterol is precipitated as the digitonide, either before or after saponification, and the color reagents are added to an acetic acid solution of the precipitate. The solution is kept at 25 C, and the absorption at 610-620 mp is determined within 27-37 minutes after addition of the reagents. A revision of this method by Sperry and Webb (1950) involved the use of a dilute alcoholic solution of 17 digitonin, rather than the aqueous solution used originally, and mixing of the color reagents prior to addition to the cholesterol solution.
Galloway and co-workers (1957) were successful in using only 0.04 ml of serum for the analyses of both free and total serum cholesterol. This micro procedure was adapted from the macro method of Sperry and Webb
(1957), and was used in this study. Thus serum samples could be obtained from the finger-tip blood, time was saved, the quantity of reagents used was reduced, and the accuracy of the method was high. METHODS AND PROCEDURE
Experimental Design
Nine healthy university students were selected to cooperate in this experiment; five women and four men. The University Health Service1 gave all the subjects a physical examination before starting the experi ment and found them to be in excellent condition. Blood pressure and pulse rate before and after exercise were checked at two week intervals and found to be normal. All subjects were maintained on self-chosen diets while eating two meals a day with no lunch or three meals a day.
In order to minimize the effect of environmental factors such as weathe;, examinations in classwork, and other factors, half the men and women were placed on each diet for period one as shown in the experimental design in Table l. Diets were reversed for each subject in the second period. Students were asked to maintain constant weight on these self chosen diets and to keep a record of all food eaten on the day before urine was collected. The content of calories, protein, fat, carbohydrate, saturated fatty acids, unsaturated fatty acids (oleic, and lenoleic), and cholesterol was then calculated from the food tables in the U.S.D.A.
Handbook number 8 (1963), and in Bowes and Church (1963).
This study was part of a larger problem which included the deter mination of the degradation products of the androgens and adrenal steroid hormones, progesterone, and also estrogens in urine and serum cholesterol levels. Two mornings a week, finger-tip blood was taken at the Food and
Nutrition Laboratory for cholesterol determination. The subjects were 19
Table 1. Experimental design
Subj. Two meals with no lunch Three meals with lunch
Period a Male Female Male Female
1 B.W. T.D. H.A. B.N.
D.P. J.A. B.E. S.R.
M.J.
H.A. B.N. B.W. T.D.
B.E. S.R. D.P. J .A.
M.J. aPeriod refers to the first 31 days. Period refers to the second 31 days. 20
then served breakfast. All food was measured for each subject at
breakfast. The mean age, weight , and height of subjects are presented
in Table 2 . Two 24 hour-urine samples were collected each week.
Additional samples of urine from the female subjects were also col~
lected on the inth, eleventh, thirteenth, fifteenth, and seventeenth
days after the first day of the menstrual period . Although the addi
tional urine samples were taken especially for estrogen analyses, they
were analyzed in this study for their content of 5~ -pregnane-30(,200(
-diol. The specific days (9-17) were chosen to observe the effect of
ovulation, which occurs usually between the thirteenth and seventeenth
day of the menstrual period, on the levels of the biochemical indices
under study; that is, on serum cholesterol level and pregnanediol .
Collection and Storage of Urine Specimens
Collection
Instructions for collecting 24- hour urine specimens were given to each subject, that is, to discard the first morning specimen on the day of collection and to collect all urine voided up to and including the first morning specimen of the next day. Containers of adequate size
(two quart bottles) labeled with name and date were given to the subjects for the 24-hour collection.
Creatinine values were used to check the accuracy of the complete ness of the total urine for the 24 -hour collection since these values have been shown to be nearly constant for a given individual. Folin's method as out-lined by Hawk et al. (1954) was used for the analysis.
Results were all within the range of creatinine values for normal subjects . 21
Table 2. Mean age, weight, and height of subjects
Sex Period No. of Age Weight Height meals year 2ounds feet inches
Male 21 166 5 11
21 166 ll
1 3 21 183 ll
21 184 11
Female 21 141
3 21 142 6
21 119
21 118 6 C!2
Although all of the urinary steroids are relatively stable com-
pounds, a preservative was added t o each bottle in which urine was to
be collected. A mixture of penicillin and streptomycin solution (1000
units of penicillin and 5 mg of s treptomycin for each 24-hour collec tion)
was chosen as the preservative to prevent both bacterial contamination
and deterioration upon frozen storage. All the specimens were then
stored at -10 C.
Analysis of Urinary 5,§ -pregnane-3ol , 200C.-diol
Determination of pregnanediol i n urine was mad e by a modification
of the method of Eberlein and Bongiovanni (1958) on thin-layer chroma-
tography. The procedure was as follows:
Equipment:
1 . Mounting board for glass plates, 22xll3 em with retaining ledges
1.8 em wide along a short and a long side. 1
2. Desaga standard applicator, Model S-II, for mechanically pro
ducang a standard thickness. 1
3. Glass plates of uniform thickness: 5x20cm and 20x20 em. 1
4. Labeling template, graduated. 1
5. Developing tank. 2
6. Racks f or glass plates : metal and wooden.
7. Oven, 110 C.
1 Brinkmann Instruments, Inc., 115 Cutter Mill Road, Great Neck, Long Island, New York.
2Arthur H. Thomas, Co., Philadelphia, Pennsylvania. 23
8. Hamilton microliter syringe 10, 20, and 50 lambda.
9 . Hot plate.
10. Heating rack for thin-layer chromatographic plates.
11. Spray bottle.
12. Filter paper, Whatman No. 1, 45x60 em sheets .
13. Pouring funnels--2~ inches in diameter.
14. Erienmeyer flasks, 50 ml.
15. Sintered glass filters, 15 ml capacity.
16. Water bath 37 C.
17. Test tube, l.SxlS em.
18. Razor blades, single edge.
19. Conical centrifuge tubes, 50, and 15 ml.
Reagent: 1 1. Silica Gel G .
2. Indicator, 4 per cent phosphomolibdic acid in absolute alcohol.
3. Iodine.
4. Acetone, reagent grade, distilled .
5. Developing solvent, choloroform:acetone:absolute ethanol
85:15:10, v/v/v .
6. Chloroform, reagent grade, distilled.
7. Nitrogen gas.
8. Benzene, reagent grade, distilled.
9. 5{3 -Pregnane-3 10. 1M, pH 4.5 acetate buffer. 11. Beta-glucuronidase, 5000 unit per ml. 3 3 Sigma Chemical Company, 3500 DeKalb St., St. Louis 18, Missouri . 24 12. lN NaOH. 13. Sulfuric acid mixture: 30 gm sodium bisulfite is cautiously added to 200 ml of concentrated sulfuric acid in a 1000 ml beaker, with stirring (fumes) under the hood, and allowed to cool. The acid is decanted into a small bottle in which it may be stored at room temper ature . Procedure : Hydrolysis. Duplicate 20 ml samples of urine were used. To 20 ml of urine in a 50 ml centrifuge tube was added 2 ml of 1M (pH 4.5) acetate buffer, and 3500 units (0.7 ml) beta-glucuronidase. Each tube was stoppered with glass stoppers to prevent any loss of the sample during hydrolysis. Hydrolyses was performed by incubating the tubes in a 37 C water bath overnight (12 to 18 hours). Two stand~rd samples, 10 microliter pregnanediol + 20 ml water + buffer + ketodase, were also incubated under standard conditions. Extraction. After hydrolysis, the samples were extracted with 10 ml benzene by the beating technique, which used a footed stirring rod for the beating. The two layers separated readily on standing and the benzene layer was transferred by means of a serum lifter to a conical 50 ml centrifuge tube. The extraction was repeated with another 10 ml of benzene and the extracts were combined. Washing and drying. The combined extracts were washed with 8 ml of lN NaOH three times, and then 8 ml of water three times. After removal of the water, the benzene extract was transferred to a conical 30 ml centrifuge tube and evaporated to dryness at 45-50 C under a gentle stream of filtered nitrogen gas. 25 Preparing t he glass plates. The glass plates were coated with Silica gel G as follows: 1 . Preparation of glass plates and spreader . The glass plates were thoroughly cleaned with a cleaning agent, rinsed well with tap water, distilled water, and air dried on a drying rack at room temperature. Just before use, the plates were f urther cleaned by washing with acetone to remove any dust or contaminating materials. 2. Preparation of the suspension of Silica gel G and filling the spreader. Silica gel G was used as the absorbent for coating the glass plates. The volume of slurry requires to coat five 20x20 em plates con tained 27 gm Silica gel G and 75 ml redistilled water in a 250 ml stoppered flask. The flask was shaken vigorously for 30 to 45 seconds, and immediately transferred to the open spreader. The spreader was held with both hands and drawn across the plates without applying much pressure. Finish this step within 4 minutes. 3. Drying the plates. The plates were left in a flat position for 10 minutes, and then dried in air overnight which gave layers that ad hered particularly well . 4. Ac tivation. The plates were placed in a drying cabinet. Time and temperature of heati ng were determined by the required activity of the layer. Heating to 110 C for 2 hours gave the correct adsorption activity . Layers with greater absorption activity can be obtained by heating for longer times at higher temperatures. 5. Storage. Since active plates became deactivated in moist air, they were stored over a desiccant in a desiccator, or in a plate cabinet . If hot plates were placed in the desiccator the stopcock must be left open, so this is provided with a short drying tube filled with 26 silica gel. The layers were thus protected as well as possible against laboratory fumes and mechanical damage. Chromatography. Just before applying the sample, impurities were removed from the plates by placing them in a glass tank, which had been saturated with chloroform. The dried sample of pregnanediol was dis solved in 50 microliter chloroform and was applied to the plate with a 20 microliter pipet at a point approximately two centimeters from the edge of the plate at right angles to the direction in which the plates were coated . A standard solution of 5fi-pregnane-3o(,20Dt-diol pre pared for thin-layer chromatography was applied to the layer as a marker for i dentification . Immediately after the sample was applied, the plate was placed in a developing tank which previously had been saturated by adding 150 ml of developing solvent. The developing solvent system used by Waldi and Munter was chloroform:acetone = 9:1 (v/v). This did not work well enough in separating the spots. By changing the ratio of the solvent system and comparing many times, it was found that using chloroform:acetone:absolute alcohol = 85:15:10 (v/v) (plate A) as the solvent was better than chloroform:acetone 90:10 (v/v) (plate B). Saturation was accomplished by lining the tank with a filter paper wick. When the solvent front had reached a pre determined mark (15 em), the plate was removed from the chamber, air dried, and quickly placed in an iodine atmosphere until the pregnane diol components became visible as yellow spots. The plate was removed from the iodine atmosphere, the area of pregnanediol was located, and the portion of silica gel G containing the pregnanediol component was immediately scraped off into an Erlenmeyer flask (50 ml) containing the eluting solvent (chloroform, Plate A. Thin-layer chr o matogram following the first isolation. Solvent sys tem- chloroform:acetone:absolute alcohol = 85:15 10 (V/V) Plate B. Thin-layer chromatogram following the first isolation. Solvent syste m-chloroform:acetone = 90:10 (V / V) Plate C. Thin-layer chromatogram following purification by re-chroma tographying. Solvent system-chloroform:acetone:absolute alcohol = 85:15:10 (V /V) I \ 28 20 ml). The silica gel G-pregnanediol mixture was filtered through a sintered glass filter into a l.5xl5 em tube. The flask was rinsed several times with solvent to insure quantitative transfer of t he pregnanediol. The solvent was evaporated j ust to dryness in a nitrogen atmosphere while the tube was held i n a warm water bath. Although thin layer chromatographic analysis of the components (pregnanediol and steroids) extracted from urine samples showed retention, it also showed tailing of the neighboring steroid classes (plate A). In order to further purify the pregnanediol , it was re-chromatographed by the thin layer procedure and isolated a second time (plate C) as described above. If a picture was to be taken of the plate after allowing spots to develop in iodine vapor , it was sprayed with 4 per cent phosphomolybidic acid in absolute ethanol under a hood. Heating on the heating rack, the pregnanediol spot appeared as blue spot. The samples which were photographes could not be used for quant itative determination, be- cause of the effect of fixing agent on pregnanediol. Spectrophotometry. To the dried sample, was added 2.6 ml of 15 per cent sodium bisulfite in sulfuric acid solution. A blank was prepared simultaneously. The: spots of the standard solution of preg nanediol were treated in the same way as the urinary extracts. The tubes were placed in boiling water bath for 4 minutes. Af ter cooling at room temperature (15 t o 20 minutes) , pour the contents i nto 3 ml silica cuvettes and read at 390, 425, and 460 pm . The intensity of the color is maximum at 425 pm. Many factors would influence the color density, such as, the length of time permitted for color development, amount of color developed by the individual pregnanediol, and the con centrations of sodium bisulfite and sulfuric acid. Therefore, Allen 's 29 correction was followed to reduce the error (allen, 1950). Calculation. The corrected absorbence was calculated by the Brown modification of the Allen ' s correction: 2 absorbence - (absorbence + absorbence ) 425 460 390 The excretion of 5beta-pregnane-3 alpha,20 alpha-dial in urine for mg per 24-hour = corrected absorbence unknown/corrected absorbence known x 24 -hour urine volume/volume of urine sample x meg of standard used x 1/1000. Recovery. The precision of the method was well established by a good recovery test. The recovery test was made in duplicate. Recovery tests were made on each subject 's urine. To a 10 ml of urine sample was added 6 meg of standard, 5/3 -pregnane-301.. ,20 at the same time with the 10 ml urine samples alone following the routine procedure. The recovery ranged from 87.6 to 103.7 per cent (Table 3). The mean percentage recovery was 96.7 with a standard error of 4.8. Most of the recover y data fell within experimental error but the results suggested that some pregnanediol might have undergone some slight chemical decomposition during the hydrolysing procedure. Preparation of Serum Samples Blood was taken from the finger-tip of subjects in the morning be- fore food was eaten. Blood was collected in l0x75 mm tubes which were sealed with Pyseal, and allowed to clot for at least half hour and then centrifuged at 2500 rpm for 30 minutes. All serum was transferred to 6x50 mm test tubes. These were stoppered with rubber stoppers, frozen, and held at -5 to -10 C until analyses were made. 30 Table 3. Recovery test of urinery pregnanediol Subjects Pregnanediol meg Recovery 10 ml urine 6;>,. standard a % + 10 ml urine BW 1.967 8.258 103.65 BN 3.994 9 . 801 98.07 TD 18.038 23.978 99.75 JA 2.790 7.699 87.59 SR 6.709 12.236 96.28 DP 2.424 7.667 91.01 MJ 7.091 12.065 92.16 HA 14.462 20.054 98.01 BE 7.312 13.763 103.38 Mean 96 .65+ 4.8b a6mcg of 5/3 -prenane-3 bMean with standard error. 31 Analysis of Serum Cholesterol The method used in this study was the micro method of Galloway et al. (1957), an adaptation of the method of Sperry and Webb (1950b). Modifications were made according to recommendations of Peseador (1959) as follows: 1. Serum samples were added to acetone-ethanol rather than the reverse order in extraction. 2. The concentration of potassium hydroxide solution used for saponification of aliquot for the total cholesterol analysis was 16.5 per cent rather than 33 per cent. The following additional changes were also made. 1. In order to more firmly pack the precipitates when using a high- speed , refrigerated centrifuge, the time for centrifuging was increased to 40 minutes at 5000 rpm. 2. Aliquots of the filtrate were pipetted directly from the 1 ml volumetric flask into 6x70 mm test tubes following the first precipita- tion and centrifugation. 3. A control serum3 sample was analyzed with each series of samples. 4. Each cholesterol digitonide precipitate was washed one extra time; the total cholesterol being washed with 200 microliters of acetone- ether prior to the ether washing; and the free cholesterol being washed a second time with 200 microliters of ether. 3clinical chemistry control serum from Hyland Laboratories, Los Angeles, California, was reconstituted from the freeze-dried state. 32 Equipment: 1 . 1 ml volumetric flasks with stoppers. 2. Pipettes: 10, 30, 40, 50, 100 , and 200 lamba pipettes. ml graduated pipettes for adding 0.4 ml of acetone-ethanol and making to volume. 3. 6x50, 6x70 mm, and 3 ml (10x75 mm) test tubes. 4. Refrigerated centrifuge or centrifuge that can be placed in refrigerator. 5. Wide-mouth pint preserving jars with lids, containing 1~ inches of sand. 6. Dark cabinet , containing a constant water bath set at 28 C. A packing box open on one side only and placed in a relatively shaded por- tion of the room will serve. 7. Stop clock. 8. Test tube racks to hmld 1 ml volumetric flasks and racks to hold 1 ml test tubes in water bath etc. 9. Beckman spectrophotometer with micro attachment as described by Bessey. 10. Incubator set at 37 C with a wire rack in pan of sand. 11. Oven at 110-115 C for drying cholesterol digitonide. 12. Vacuum line. 13. Handi-tool with a small flattened nail in the end (Chicago Wheel Mfg. Co .), used to buzz the solution. 14. Syringe micro buret model no. SB2"· ~ 4Micro-Metric Instrument Co., Cleveland, Ohio. 33 Reagent: 1. Acetone-ethanol mixture, 1 :1 : One volume of redistilled acetone is mixed with one volume of absolute alcohol, redistilled. 2 . Ether, peroxide-free. Bakers absolute ether is peroxide-free. 3. Ace t one-ethe r mixture , 1:2 : One volume of redistilled acetone and 2 volumes of peroxide-free ether. 4. Digitonin solution, 0.5 per cent : Dissolve 500 mg of digitonin i n 100 ml of 50 per cent alcbhol (55 ml of 95 per cent alcohol and 45 ml of distilled water). 5. Potassium hydroxide solution , 16.5 per cent : Dissolve 1.65 gm of pure po tassium hydroxide in 10 ml of freshly boiled distilled water. It is important to have potassium hydroxide carbonate free. It is best to make a fresh solution about once every 2 weeks. 6. Phenolphthalein solution, 1 per cent in alcohol 0 . 3 gm phenolphthalein dissolved in 30 ml of 95 per cent ethyl alcohol . 7. Acetic acid solution, 10 per cent Dilute 2.5 ml glacial acetic acid to 25 ml with distilled water. 8. Acetic acid, glacial (ASC). 9 . Acetic anhydride (ASC), 99-100 per cent chloride free. If sediment or color is present, redistill. Impure acetic anhydride may make the standard run high. 10. Sulfuric acid, concentrated, C. P. 11. Stock standard solution of cholesterol in glacial acetic acid, 100 mg per 100 ml Weigh exactly 100 mg of cholesterol into 50 ml beaker. Dissolve with glacial acetic acid and pour with stirring rod into a volumetric flask and dilute to 100 ml. Heat over hot water if cholesterol is hard to dissolve. 34 12. Working standard solution of cholesterol in glacial acetic acid , 0.1 mg per ml: Dilute 10 ml of the s tock standard to 100 ml with glacial acetic acid in volumetric flask. Procedure: Extraction and sampling . The frozen serum sample was warmed to room temperature and mixed gently to insure a uniform sample. Forty microliters of serum were delivered with a 40 microliter pipette into a 1 ml volumetric flask containing 0.4 ml of acetone-ethanol (1:1). Each flask was buzzed immediately until the precipitate was divided into fine paricles. The solvent was brought to a boil by placi ng each flask (held with tongs) in a beaker of boiling water. The flask was swirled gently in the boiling water and kept boiling until t he solvent was reduced to half of its initial volume. The flask was removed from the boiling water occasionally in order to prevent vigorous bumping . The flask was cooled to room temperature at once by placing it in cold water, and the solution was made up to 1 ml with acetone-ethanol (1:1), stopper ed and shaken well . The solution was centrifuged at 5000 rpm for 40 minutes at 0 C so that the precipitate was packed sufficient to allow aliquots to be measured directly from the flask. Suplicate aliquots of 100 microliters of the sample were pipetted i nto 6x70 mm tubes for the determi nation of total cholesterol; and 200 microliter aliquots for free cholesterol determination. Precipitation of total cholesterol . Using a syringe microburet, 5 micr oliters of 16.5 per cent potassium hydroxide solution was added to the 100 microliter aliquots. Each tube was buzzed until it was well emulsified. fitted with a rubber cap, and placed in a shallow pan 35 containing a layer of one and one-half inches of sand, and then in an incubator at 37 C for 30 minutes. After cooling to room temperature, 120 microliters of acetone--ethanol were added to make approximately 200 microliters of solution. Five microliters of phenophthalein were added as an indicator and then 30 microliters of 10 per cent acetic acid were added to make the contents s lightly acid in reac tion. If a slight color remained after buzzing, an additional 5 microliters of the acid were added to insure excess acidity. Then 100 microliters of digitonin solution was added and the tubes were buzzed well. The tubes were fitted with rubber caps and placed in cover ed wide-mouth pint jars with one and one-hal f inches of sand and allowed to stand over night at room rempera ture. Duplicate blanks of 100 microliters of acetone-ethanol were treated in the same manner. Precipitation of free cholesterol. Five microliters of 10 per cent acetic acid and 100 microliters digi tonin were added to the 200 micro liter aliquots . Each tube was buzzed well, stoppered and stored over night in the same manner as for total cholesterol. Suplicate blanks of 200 microl i ters of acetone-ethanol were prepared by the same procedure. Washing of the cholesterol digitonide precipitates. The next day the rubber caps were removed. Each tube was tapped well with the finger to free any preci pitate which adhered to the wall near the surface of the liquid, and the tubes were centrifuged at 5000 rpm at 0 C for 30 minutes. The supernatant was drawn off with a fine tipped transfer pipette using very slow suctin. The walls of the tube were washed well with 200 microliters of acetone-ether (1:2) from a synringe microburet. The tubes were buzzed well to wash the precipitate and then centrifuged for 20 minutes . The supernatant was removed as above. 36 For the total cholest erol determination, the precipitates were washed once more in the same manner with ether. The precipitates for the free cholesterol determination were washed twice more with ether. After washing, the tubes were placed in a pint jar and heated in the hot water bath for the evaporation of all remaining ether from the precipi tates . The samples may be stored several days at this stage. Color development. For the development of the color, the cholesterol digitonide was dried in an oven at 100-110 C for 30 minutes. Fifty micro liters of pure glacial acetic acid were added while the tube was in the hot sand bath. After 2 or 3 minutes the tube was rotated slightly on its side to make sure the acid came in contact with all the precipitate. The tubes were tapped with the finger about 5 times and placed in a wire rack. This process of adding and mixing was continued until all the tubes were in the rack in the order of reading. A check tube containing 50 micro liter of glacial acetic acid was prepared to use in setting the Beckman Spectrophotometer at zero. Three tubes of standard solution (5 micro gram of cholesterol per 50 microliters) were equally spaced in the wire rack, one at the beginning, one in the middle, and one at the end of the series of tubes, and the rack of tubes was placed in a 28 C water bath in a dark cabinet. The Liebermann-Burchard reagent was prepared as follows: 0.5 ml of concentrated sulfuric acid was added drop by drop to the 10 ml ice-cold acetic anhydride and shaken vigorously for a few moments and placed in an ice bath for 10 minutes before being used. To each tube was added 100 microliters of this reagent from a syringe microburet in a group of three tubes every three minutes . The tubes 37 were buzzed and replaced in the wat er bath for col or development. At exactly 30 minutes after the addition of the color reagent, optical densities of each group were measured at a slit width of 0.44 mm and a wave length of 625 um on the Beckman DU Spectrophotometer against a r eference solution of glacial acetic acid. The cuvets were not washed between readings, but all liquid was drawn out with a transfer pi pet by the use of suc tion. One cuvet only was used for check solution, one only for reagent blanks, one only ~or the standard, and 3 other cuvets for samples. Calculation The calculation is as follows: Free or total cholesterol in serum (mg choles t erol per 100 ml serum) ~ Density of sample - density of reagent blank x mg cholesterol in Density of standard 100 standard X------Volume of serum in aliquots Recovery The recovery test was made on each subject's specimen in duplicate. To a 30 microliter of serum sample was added 10 meg of serum standard, which was run at the same time with the 30 microliter of serum sample alone following the routine procedure. The recovery ranged from 88 to 98 per cent for the cholesterol with hydrolysis; and from 89 to 98 per cent for the cholesterol without hydrolysis (Table 4). The mean per- centage recoveries were 93 with a standard error of 3.1, and 92 with a standard error of 3. 5 for the cholesterol with and without hydrolysis respectively. 38 Table 4. Recovery t est of serum cholesterol Sub Cholesterol after hydrolysis Cho lesterol without hydrolysis ject 307\ serum 307\ serum + Recovery 301\ serum 30A. serum + Recovery lOmcg stand. lOmcg stand. mg mg % mg mg % HY .0623 .0175 92.0 .0146 .0241 95 DP .06175 .0717 96.5 .0147 .0243 96 MJ .0634 .0732 98.0 .0156 .0251 95 SR .0591 . 0688 97.0 .0139 .0234 95 BW .0607 .0700 93.0 .0149 .0241 92 JA .0787 .0877 90.0 .0186 .0275 89 TD .0712 .0803 91.0 .0170 .0260 90 HA .0619 .07125 94.5 .0148 .0238 90 BE .0691 .0779 88.0 .0156 .0245 89 BN .0704 .0794 90.0 .0167 .0254 87 HY .0609 .0701 92.0 .0142 .0240 98 Mean 93+3.la 92+3.5a aMean with standard error RESULTS AND DISCUSSION Urinary Bregnanediol Level No significant differences in the amount of pregnanediol excreted in the urine were observed between sampling dates; sex; and number of meals (Table 5). Values for the men dad not vary much from day to day while the excretion of pregnanediol of women increased three fold or more during the second half of the menstrual cycle. Mean concentrations of urinary pregnanediol excretion values are presented in Table 6, individual data in Tables 18, 19, and 20. The overall means in this study for urinary pregnanediol output per day were found to be 0.77 mg for men; for women, 0.81 for days 1- 14; 1.77 for days 15- 28 of the menstrual cycle (Table 6). Thus, the excretion values for the men were similar to those for the women for the first 14 days 6f the menstrual cycle, and half that of the luteal phase or latter half of the cycle . The range of values found for the excretion of pregnanediol by four normal men (mean age 21) was 0.15 to 2.46 mg per day, and for five normal women (mean age 21) 0.14 to 1.85 for the first half of the men strual cycle and 0.20 to 5.76 for the latter half. However, the low values of 0 .20 and 0.32 occurring on days 15 and 17 were shown by a subject whose cycle was longer than normal or 40 days. Her values showed a definite rise after day 17. The range of values fell within the acceptable range of normal adults given by the other investigators. In Table 7 is found a comparison of the range of normal values determined by this method with those reported previously. 40 Table 5. Summary of mean squares obtained by analysis of variance of data on urinary pregnanediol. Source of variance d. f. Mean squares Sampling dates ll 1. 391733 Meals 1 0.919917 Sampling dates vs meals 10 1. 438915 Sex 1 1.262629 Sex vs sampling dates 11 1. 527742 Sex vs meals 1.611501 Experimental error 72 1.647265 Sampling error 96 0.011286 Total 191 0. 801379 41 Table 6. Mean values of pregnanediol in urine of subjects consuming two meals and three meals per day. Pregnanediol mg 12er 24 hour urine Subject Day of study 2 meals 3 meals Mean Males 3-4 0.80 1.03 0.92 10-11 0. 71 0.97 0.84 17-18 0. 74 0.83 0.79 24-25 0.53 0.51 0.52 30-31 0.68 0.94 0.81 Mean 0.69 0.86 o. 77 Females 3-4 1.32 1. 59 1.46 10-11 1. 73 0.63 1.18 17-18 0.95 1.56 1.25 24-25 1.67 1.91 1. 79 30-31 2. 84 1.51 2.17 Mean 1. 70 1.44 1.57 Females Day of menstrual no . of cycle subjects 4;2 1-3 0.85 0.98 0.92 4a 4-6 0.91 5 9 0.66 0.78 0. 72 5 11 0.61 0.89 0.75 5 13 0.67 0.90 0.78 5 15 1.08 1.43 1.25 5 17 1. 28 1.08 1.18 2a 19-20 2 . 98 sa 23- 25 1.93 3a 26-28 1.53 aNumber of subjects per meal too limited to list values separately. Table 7. Urinary pregnanediol values of normal humans as reported in the literature Sex No. of Range Average Method Investigator Mark Subjects F 9 0.31- 3 . 66 paper Eberlein et al M 1 0.45-0.56 chromatography (1958) F 4 0. 78-1.50 1.12 column Klepper et al . . Proli ferative phase 2 2.4 -4.2 3.3 chromatography (1955) luteal phase 5 0.28-0.86 0.63 post-menopausal M 9 0. 38-1.42 0 . 92 F 1 Grollman Mg/1000 ml, pre- (1941) ovulatory phase 6 late post ovulatory phase F 1. 79-5.97 Cantarow et al. Luteal phase (1962) M 0 . 1 -0.8 Astwood et al. Mg/1000 ml urine (1941) M 4 0.15-2.46 0. 77 thin-layer This study chromatography F 0.14-1.85 0.87 Proliferative phase 0.20-5.76 1.77 Luteal phase .1>- N 43 Effect of menstrual cycle The mean pregnanediol values, although similar from days 9 to 13, showed a slight tendency to rise which was followed by a sharp increase on day 15 and a further sharp increase from day 17 to its peak on days 19 - 20. The levels then decreased rapidly t o the beginning of the next period (Table 6, and Figure 2). The variation among these subjects is shown in Figure 3. Because of the variation between subjects, it would be highly desirable to obtain similar data daily on more subjects to observe the significant of the effect of the menstrual cycle on this metabolite . Individual data of the excretion of pregnanediol in the urine of the women on specified dates of the menstru~l cycl e are presented in Figures 4 to 8. The excretion of pregnanediol rose in the expected fashion during the second half of the menstrual cycle, except for one subject (J . A.) during period 1, and precipitously fell prior to the on set of menstruation. Smaller amounts of the steroid, possibly derived from the metabolism of adrenal progesterone, were detectable during the first half of the cycle. The amount of pregnanediol excreted by these subjects late in the cycle was of the order of magnitude reported by other investigators studying normal female adults (Eberlein et al, 1958; Klapper et al., 1955; Grollman, 1941; Venning and Browne, 1937; Cantarow et a l ., 1962). Effect of two meals versus three meals per day Although male subjects while eating two meals a day showed a gradual decrease in excretion of pregnanediol between days 3 and 24 of the study period, the greatest decrease occurred between days 17 and 24 (Table and Figures 9 and 10). These excretion values returned t o the val ue tdpregnanedrol 3.0 220 8 ()Q O cholesterol ~ ... ~ g 2.5 210 "'>-' ~ n II ::r 0 .,.."' ,>-' "... 2.0 ~ 200 ;:l ,"'rt ... .., ;:l [/ 0 0 >-' .c ll ., I 1.5 190 ,.., "" [I "' >-' ... 0 ~ 0 "'p.. [I ~ .-< 1. 0 180 ,_.8 ...,0 v ., II , 17 ..,"' "' v ~ c: <11 1/ " 0.5 v 170 8 1/ ...""" / "'p.. v ~ / v ~ ~ abO v [I [I c 0 160 1-3 4-6 11 13 15 17 19- 20 23- 25 26-28 Days of mens trual cycle Figure 2. The effect of day of mens trual cycle on excr etion values of pregnanediol and t he level s of serum cholesterol 5.0 4. 4. ~ 4 -- - · 3.5 .,.._ - -¥ 3.0 ---- QJ .,;" 2.5 " " .c0" 2 . .,.I N ~ ' QJ "0.. ' .... " ' 0 ' .,.,; 1. " QJ \ " SR "' 0.5 \..MJ """QJ "0.. ""8 0 1-3 4-6 11 13 15 17 19-20 23-25 26-28 Figure 3 . The effect of day of menstrual cycle on mean excretion values of pregnanediol of 5 subjects ~ c h o 1 e sterol a 6 2 00 ""rt ~ pregnanediol 0 .,rt .... n 1 90 ::r ....0 11> fJl rt ., 11> ..... 180 ... " 4 ....0 " "'11> "0 ... .e" 170 ...... ,.I 0 "' 0 ., ....a "p. 160 fJl ..-< 2 0 "'...... , § "'d 150 .,a"' "p. 00 [ ~ i! 100 13 l ~ ~ ~ J ~~ 10 15 20 25 10 15 20 25 30 Days of menstrual cycle Figure 4. Urinar y pregnanediol excretion values and serum cholest erol levels by day of menstrual cycle for subject S.R. 5.0 40 Jl ~ cholesterol ,.n 4 .5 ....0 ~ pregnanediol f2 30 "'~ 4 .0 "''1 ....0 f2 20 -g 3.5 ".... 0 0 3.0 .2 10 ....a "' QJ "' .,...~ 2.5 §" ... ..2 00 ..." 2.0 0" -"I 1 90 N""" ... 1.5 "'0. 1 80 .... .,...0 1.0 ""'QJ ~ 1 70 :;b"' 0 . 5 ,.. "'0. ~0 ~ ~ ~ 1 60 10 15 20 25 30 35 40 10 15 20 25 30 Days of menstrual cycle Figure 5 . Urinary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycl e for subject T. D. 4.0 I r ~cholesterol s ()Q 3.5 ~pregnanediol 300 N 0 .,N .... 3.0 280 ::r" ....0 "'~ 2.5 ..."' 260 ....0 "0 ..."' ~ 2.0 .... 240 ...... 0 ~ 0 ... s ~ .... ] 1.5 Ill 220 I ..."' ~ sc: ... ~ 0 2 20 ° a ~ rn QJ ..."' .;; 2.0 2 00 § >-< ::J >-< ::J 5 _g 1. 1 80 I "'""' >-< QJ P. 1. 0 1 60 .-< 0 ·r< ""'QJ .:: OJ 5 .:: 40 "bO QJ >-< P. 0 bO ~ 20 s LlJ jlJ :J' LU _J._J LU Days of menstrual cycle Figure 7. Urinary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycle for subject B.N. ~ cholesterol 3.5 3 3 00 I)Q ~ pregnanediol ,.., .,,..,0 3.0 2 80 ..... "::r .....0 CD 2.5 2 60 00,.., CD "0..... QJ '0 -;:" 2.0 2 40 CD :;l "...... 0 :;l 0 0 f 1.5 lz 20 .....3 ""N.. QJ p. 1 . 0 00 rl 0 ..-< ""''" § 0. 5 80 OD " Figure 8 . Uri nary pregnanediol excretion values and serum cholesterol levels by day of menstrual cycle for s ubject M.J. 02meals E1 3 meals Figure 9 . The effect of numbe r of meals on mean excretion values of urinary pregnanediol for male subjects "'0 2.5 2 meals 3 meals 2.0 BE 1.5 1'-, I ' I '-. ' -,... ' 1.0 ' ' ~ DP - ~ --~ ... _ __ __ ,.. . ~ BW .'\. / ' HA ' 0 . 5 BE 3- 4 10-11 17- 18 24 - 25 30- 31 3- 4 10- 11 17- 18 24 - 25 30- 31 Days of study Figure 10 . The effect of number of meals by day of test period on pregnanediol excreti on values f or male subject s 52 obtained on day 17 during the last 7 days. Men who had three meals a day, had a similar but more striking pattern than those on two meals a day. The pregnanediol values for men during the 31 days were somewhat higher during the three meals test period than for the two meals test period. The lowest value of pregnanediol excreted by the men while eating two or three meals daily occurred on day 24 of the test period. However, the wide variation shown by one subject in each group probably influenced the means (DP on two meals and BE on three meals). More subjects are needed to determine whether number of meals influence pregnanediol excretion. This wide variation between individual ex cretion values for men by number of meal, and day of test period is shown in Figure 10. The true significance of changes in excretion values for women can only be observed when day of menstrual cycle is considered. Much less variation in values was shown by the men than by the women . The women with lunch showed slightly higher pregnanediol values than those without lunch which is similar to findings for the men (Tamlesc6 and 19! Figure 11). The values for women without lunch by day of menstrual cycle showed a marked increase to days 19-20, and then dropped rapidly t o days 1-3 of the next period. During the regime with lunch, a gradual increase of excretion occurred from day 9 to day 15, and was followed by a short drop to day 17, and a marked increase to its high peak on days 19-20. The curve then dropped ~harply to days 1-3 of the next period. However, these apparent differences in excretion because of the effect of the number of meals per day were not statistical significant. Examples of individual variation in excretion values for four women by number of meals, and day of menstrual cycle are illustrated by Figure 12. 4 0 2 meals ~ 3 meals 3 1 0 Days of menstrual cycle Figur e 11. The effect of number of meals on mean excretion values of urinary pregnanediol for fema le subjects 4 .5 JA /( 2 meals 3 meals I D 4.0 I 3,5 I I I I 3.Q I TD 2 .5 I I MJ 2.0 I I 1 . \ I \ ... _.-; \JA O·L---~--~----4---~----~--4---~---L----~--4---~----~--~--~ 10 15 20 25 30 35 5'* 10 15 20 25 30 V> Figure 12 The effect of number of ;.eals bv dav of menstrual cycle on pregnanediol excretion val ues "" 55 Serum Cholesterol Level Statistical analysis of the data (Tables 8 and 9) showed that significant differences existed between sampling dates for the free and total serum cholesterol values for all subjects and the per cent of free in total cholesterol or for the female subj ects alone. The serum cholesterol values varied from day to day in each subject mainly due to the complex mechanisms involved in controlling cholesterol in the body (Cook, 1958). For the f emales, the menstrual cycle would be re sponsible for much of this day to day variation. No sex differences were observed for both the free and total serum cholester ol on these subjects with an average age 21. This was in agreement with many other authors (Keys et al., 1950, 1952; Kornerup, 1950; Jones , 1951; Adlersberg , 1956; Lewis, 1957). The interaction of sampling dates by meals was significant at the one per cent probability for both the free and total serum cholesterol . In Table 10 is presented the percentage distribution of the 9 sub jects by sex within various ranges of serum total cholesterol values . Most of the values (69 per cent) were within the range of 170 to 229 mg per 100 ml of serum when the individual determinations were tabulated. Mean values for the entire study for each s ubject showed 78 per cent within this range. The individual values are shown in Table 20 . Mean free cholesterol values were 54 .5 for men and 53.6 mg per 100 ml of serum for women. The range for the per cent of free in total cholesterol was found t o be from 20.5 to 32.4 per cent. This percentage seemed to be constant, although the total and free cholesterol varied considerably. · The mean values of the per cent of free in total choles terol were found to be 25.6 and 26.4 per cent for men and women. The Table 8. Summary of mean squares obtained by analysis of variance of data on serum cholesterol of male and female subjects. Mean squares Percentage of free Source of variance d . f. Total cholesterol Free cholesterol in total cholesterol Dates 17 ** 8127.77 377.4424 ( .01)** 18 .6836 Meals 95.68 25.6806 59.5868* Dates vs. meals 16 8629. 78** 399.4275** 16.1272 Sex 572.34 3 . 0422 7.2200 Sex vs. dates 17 1615.52 252.8305 26.7210 * Sex vs. meals 1 360.02 2.0672 12.0050 Experimental error 108 1619.25 166.7156 12 . 4665 Sampling error 144 5.03 0.6525 0.0541 Total 287 1190.98 100. 4074 7.4330 *significant at 5 per cent probability **Significant at 1 per cent probabi~ity Table 9. Summary of mean squares obtained by analysis of variance of data on serum cholesterol of female subjects Mean squares Percentage of f ree Source of variance d. f. Total cholesterol Free cholesterol in total cholestrol Dates 17 3818.15** 344.2753* 12 . 605* Meals 1 649.80 164 . 3556 111.5494 ** Dates vs . meals 16 4016 . 17** 355.5203 * 6.4211 Subjects 898.59 140.6115 17.3065* Experimental error 68 923.03 166.9130 6.6917 Sampling error 90 550.83 0.9467 6.7946 Total 179 1010.30 99.7229 7.5423 =~ig n ificant at 5 per cent probability Significant at 1 per cent probability 58 Table 10. Percentage distribution of nine subjects within various ranges of serum total cholesterol values Range of cholesterol Numb era Percentage Number mg per 100 ml serum M F M & F M F 130-139 1.2 140-149 0 0 . 6 150-159 l 5.0 160-169 1 1.9 170-179 8.7 180-189 ll 10. 6 l 190-199 8 20 17.4 200-209 10 8. 7 210-219 13 13.7 220-229 8 10.5 1 230-239 7 . 4 1 1 240-249 3 5 .0 250- 259 3 5.0 260-269 1.9 270-279 0 1.2 280-289 0 0.6 290- 299 0 0.6 aNumber of individual subjects whose cholesterol values for the different days of the study fell within each range (71 determinations for the 4 men and 90 for the 5 women) bMean cholesterol values for entire study for each subject 59 data for serum free cholesterol levels are presented in Tables 11 and 12; individual data in Table 21. The overal means for serum total cholesterol levels were found to be 214 ± 33 and 204 ± 31 mg per 100 ml of serum for men and women, respectively. These results were in agreement with the values reported by other workers. Man and Peters (1933) observed that 12 subjects had total serum cholesterol values about 207 ± 29 mg. Observation on 118 subjects gave result of 152 ± 24 mg in Boyd's 1937 study . Gartler and Garn (1950) found that the total serum cholesterol of 146 men were 224 . 4 ± 35 mg per 100 ml. It was shown by Bohle and Biegler (1958) that serum cholesterol values for 513 men and women were 205 mg per 100 ml. The range of values shown by the 71 and 90 cholesterol determinations for the men and women, respectively, were 1.37 to 280 and 130 to 278 mg per 100 ml serum. The subjects had similar ranges of cholesterol values when eating 2 or 3 meals a day (Table ' l3)~ These ranges were somewhat wider than those found to be normal for adults by other investigators but this study involved many more determinations per subject. When only mean values for each subject were considered, the r anges were 196 to 225 and 179 to 228 for the men and women, respectively. The range of serum cholersterol levels between subjects was reported to be 150 to 280; 150 to 250 ; 138 to 296 mg per 100 ml by Krupp et al. (1960); West and Todd (1962); Galloway e t al. (1957), respectively. Although there are great differences on so called normal serum cholesterol values, the generally accepted range in adults is between 160 and 250 mg per 100 ml (Wilcox et al., 1962). A more recent suggested desirable cholesterol value for persons with coronar y disease risk, is well under 200 mg or less (Stamler et al., 1963). 60 Table 11. Mean serum cholesteral concentration by sex and meals Sex No. of No. of Total cholesterala Free cholesterala Free in total meals subjects cholesterol a mg/100 ml mg/100 ml % M 206 ± 29 52.0 + 7.6 25~3 ' ± ' 2:4 221 ± 37 57 .0 ±13.1 25.8 ± 2.4 F 2 5 205 + 28 52.7 + 9.8 25.7 + 2.7 3 5 202 ± 35 54 . 5 ± 9.1 27.2 ± 2.3 aMean with standard error 61 Table 12. Mean values of total and free cholesterol in serum of subjects consuming two and three meals per day Subjects Day of Total cholesteral Free cholesterol study 2 meals 3 meals 2 meals 3 meals Males 3-4 195 189 51.4 51.6 10-11 214 126 55.0 53.8 17-18 207 246 50.1 61.9 24-25 207 244 48.6 69.1 30-31 210 203 52.7 52.1 Mean 207 220 52.7 52.1 Females 3-4 187 197 46.9 56.3 10-11 211 206 55.2 55.3 17-18 224 208 56.4 54.4 24-25 217 199 56.4 57.0 30-31 196 206 51.5 52.8 Mean 207 203 53.3 55.2 Day of menstrual cycle 1-3 181 221 4-5 197 188 6-7 264 198 8-9 210 251 10-11 203 198 13 247 202 15 184 230 17 227 204 19-20 206 187 22-23 199 204 24-25 213 208 26-28 177 29-30 204 196 62 Table 13 . The range of serum total cholesteral values for individual subjects by number of meals Men No. of Total cholesteral Women No. of Total cholesteral meals range meals range BW 186-267 TD 177-225 210-242 175-211 DP 2 195-240 JA 181-261 3 179-280 195-254 HA 137-206 MJ 185-278 171-227 169-274 BE 2 152-236 BN 150-261 3 171-235 130-231 SR 2 165-194 3 151-220 Effect of menstrual cycle The effect of menstrual cycle on mean serum cholesterol level is shown in Figure 2 and Table 14. Higher values were observed from days 6 to 17. Soon after ovulation, values decreased and were lower for days 18 to 28 with a slight increase to day 5 of the next period . Thus, concentration of serum cholesterol was found somewhat lower during the second half of the menstrual period (luteal phase) than was the first half of the cycle (foll1cular phase). These results were just the reverse to the pregnanediol excretion values. Since the blood samples were not collected on as many specific days of menstrual cycle for each subject as for the urine specimens, no definite conclusions can be made. Results of this study indicate an agreement with other workers (Okey et al., 1927, 1933; Kritchevsky, 1958) that the menstrual cycle does affect the concentration of cholesterol in the blood. It would be desirable to obtain similar data on each day of the menstrual cycle for more subjects to observe the effect of the menstrual cycle on serum cholesterol level. Effect of two meals versus three meals per day The number of meals (2 vs 3) did not have a significant effect on the free, and total serum cholesterol values as shown by statistical analysis (Tables 8 and 11) . These men on self- selected diets showed considerable variation in total serum cholesterol level between certain days when 2 meals were eaten (Figure 13 and Table 12). Marked differences were noted on 3 meals per day for men. The reason for this variation was not apparent . The male subjects had slightly higher serum cholesterol values while on 3 64 Table 14 . The effect of menstrual cycle on the concentration of the serum cholesteral Days of menstrual l-3 4-6 11 13 15 17 19- 20 23- 25 26-28 29-30 cycle Mean serum cholesteral 192 194 223 209 213 207 217 198 214 177 199 mg/lOOml serum No . of subjects 4 4 4 260 o2 meals ~ 3 meals 240 [I 220 s v ...::> "'(f) 200 ~ .... / s v II 0 v 0 t;;: v .-< 180 v ... v II "'p. ~ ll bO v s 160 ~ ~ ~ 3 4 10 11 17 18 2 2 3 31 Days of study Figure 13 . Mean total cholesterol values for men with or without lunch by days of study 66 meals a day than on 2 meals a day. Women consuming 2 meals a day showed somewhat greater variation in cholesterol values from day to day of the menstrual cycle than on 3 meals a day (Table 12 and Figure 14) . The female subjects had about the same serum cholesterol values during the 2 test periods . The variation in serum cholesterol level for the individual men and women by number of meals and day of test period is illustrated in Figures 15 and 16. The variation between indiv idual subjects of the same sex was also great. The results of this study did not indicate any definite evidence that the number of meals , 2 versus 3, eaten per day had an effect on serum cholesterol values. Effect of dietary factors During the experimental periods, the subjects kept a record of the food eaten each day before specimens were taken. In Table 15 is pre sented the daily intake as calculated of calories , proteins, carbohy drates, fats, fatty acids (saturated and unsaturated), and cholesterol. The male subjects were consuming almost the same numbers of calories when they ate two meals a day (1.4 per cent less) as when they ate three meals a day. In case of female subjects, the caloric intake was 6.5 per cent less when they had only two meals a day. However, this dif f~rence in the caloric intake is not great enough to effect this s tudy. The total caloric intakes of the male subjects compared favorably with the NRC Recommended Daily Dietary Allowance for men aged 25 , while the women ranked a little lower than the recommended allowance (2107 vs 2300). The percentage of calories consumed as prot ein was 13, as fat, 45, and as carbohydrate, 42. These percentages were almost the 270 o 2 meal s ~ 3 meals 250 230 3 QJ "U) 210 ..... s 0 0 ..... 190 "QJ "'Oil s 170 r ~ r ~ n 1-3 4- 5 6-7 8- 9 10- 11 13 15 _l] 19 ~ 20v22 - 23 ~. 26-28 24-25 29 - 30 Days of menstrual cycle Figure 14. Mean total cholesterol values for women with or without lunch by days of menstrual cycle 2 meals 3 meals 275 I I w I " 250 DP } / I I 225 I BW I ' / ' '...,: 200 § .,... ' I BE ..-<"' ',( 8 175 0 HA 0 ..-< BE .,... p., e00 150 3- 4 10- 11 17-18 24- 25 30- 31 3- 4 10- 11 17- 18 24-25 30- 31 Cays of Study ~ Figure 15 . The effect of number of meals by days of study on total serum cholesterol values 00 for men 2 meals 3 meals 280 260 _..JA / ,_....__ -- JA 240 I --...-- / I / / 220 I I / I / ' 200 I """' I ;,.' TD I ' If"! 180 ttt-- - I I: I ~' I "' \"'-sa //\, '\ ,, I -J / "i .lr..BN l' 5 10 15 20 25 30 35 10 15 20 25 30 Days of menstrual cycle Figure 16 . The effect of number of meals by days of mens trual cycle on total serum cholesterol values Table 15. Mean food intake of the subjects Sub No. of Order of Calories Weight Proteins Fats Carbohy- Fatty acids Chole j ect meal s diet drates Satu. Unsaturated sterol t:>leic ll.ino. Women period lbs gm gm gm gm gm gm mg B.N. 2nd 1824 103 58 87 203 38 32 5.0 557 J.A. lst 2080 139 70 100 224 40 36 9 . 8 526 S.R. 2nd 1883 132 61 88 212 28 32 7.3 485 M.J. lst 2016 147 73 105 194 40 36 9 . 2 638 T.D. 1st 2052 136 69 84 254 30 37 7. 2 518 mean 1971 132 66 ----g) 217 35 35~ """"54i3 percentage of calories 13 42 44 18 18 0.4 B.N. 3 1st 2148 104 72 100 239 37 41 9.0 563 J .A. 3 2nd 2158 142 76 99 241 37 38 9.6 438 S.R. 3 1st 1921 134 56 92 216 26 29 5. 9 408 M.J. 3 2nd 2187 146 76 104 236 38 44 10.0 609 T.D. 3 2nd 2123 137 69 106 223 33 37 8.8 458 mean 2107 133 --=j(J 160 ~ 34 38 ----s:7 495 percentage of calories 13 43 44 17 19 0 . 4 Men B.W. 1st 3361 166 109 168 353 73 63 11.1 1057 H.A. 2nd 2924 150 91 154 292 53 58 12.3 1150 D.P. 1st 3156 166 118 147 339 54 55 13.2 1069 B.E . 2nd ~ 214 122 187 300 80 _J.Q__~ 1287 mean 3205 ~ liO ~ 321 65 61 11.5 1141 p ercentage of calories 14 46 40 20 19 0.4 B. W. 3 2nd 3216 164 121 168 306 63 70 16.4 1257 H. A . 3 lst 3266 150 106 170 327 61 68 11.7 1090 D. P . 3 2nd 2945 167 110 147 295 62 60 11 . 8 1853 B.E . 1st 3574 213 134 191 330 76 71 14.3 1249 mean 3205 ~ 118 169 ---w;---- 66 6713:5 lii2 percentage o f calories 15 47 39 20 .2 19 . 4 0.4 71 same for two or three meals. This intake of protein compares favorably with recommendations of nutritionists to use 10 to 15 per cent of the calories as protein for good nutrition . The fat intake would be a high intake as a medium level is considered to be approximately 35 per cent. Likewise, the carbohydrate intake is somewhat low as it should be 50 per cent of the calories or near that figure. There was no marked difference of saturated fatty acids , unsaturated fatty acids, and cholesterol intake by male and female subjects between the two meals and three meals a day regimes. Values for the two and thr e~ meal periods for the male and female subjects were 65 vs 66 and 35 vs 34 gm of saturated fatty acids, respectively, which accounted for 20 vs 20.2 and 18 vs 17 per cent of the total calories; 61 vs 67 and 35 vs 38 gm oleic acid (19 vs 19.4 and 18 vs 19 per cent of total calories); 11.5 vs 13.5 and 7.7 vs 8.7 gm of linoleic acid (0.4 vs 0.4 and 0.4 vs 0.4 per cent of total calories); and 1141 vs 1112 and 548 vs 495 mg of cholesterol per day. The ratio of polyunsaturated to saturated fatty acids was 0.23 which indicated a high percentage of saturated fat. The intake of saturated fatty acids, poly-unsaturated fatty acids (linoleic acid), and cholesterol were somewhat higher than the amount recommended by j tamler et al. (1963). Their recommendation to prevent altherosclerotic coronary heart disease, was to use diets that contain moderate intakes of the following nutrients, thus decreasing the quantity in the usual American diet: 1. Calories, 1700 2. Total fat, 30 instead of 45% 3. Saturated fatty acids, 20 instead of 44 g; 10 instead of 1 7~ 72 4. Dietary cholesterol , 300 instead of 600 mg 5 . P/S ratio, approaching 1 . 0 rather than 0.3 to 0.4. They recommended that the diets also be high in good quality protein, vitamins, and minerals. Relationships Between Serum Cholesterol and Degradation Products of Androgens and Progesterone The relationships of serum cholest erol and degradation products of androgens and progesterone, that is, neutral 17-ketosteroid and preg- nanediol are shown in Figures 17 a nd 18 and Table 16. The information on urinary 17-ketosteroids was taken f rom data in Yu's thesis (1964) on the same subjects. The male subjects had the higher cholesterol level, but lower neutral 17-ketosteroid values between days 10 to 25. In contrast, they had the lower c holes t erol level, but higher neutral 17-ketosteroid values on days 3 t o 4 and 30 to 31, at the beginning and end of study. The female subjects had higher cholesterol levels , but lower pregnanediol values on days 9 to 13 of the menstrual cycle. They had the lower cholesterol level and higher urinary pregnanediol values on days 19 to 28. In conclusion, the concentration of serum cholesterol of men and women had a reciprocal relationship to their sex hormones , neutral 17-ketosteroid in males and pregnanediol in females. This result was in agreement with the knowledge that cholesterol served as the precursor of sex hormones and adrenal cortical s t eroids . 73 12 ~ pregnanediol 0 cholesterol 11 ~ neutral-17-ketosteroid 10 El ) OQ n ::r ....0 "'Ul rt ~ "' 8 240 8.... Q) ~ .... "" " ""' " 230 :::; " 0 "0 El .<: .... I "'" 6 220 ~ "' " "Q) ~ " I'J "".... 210 0 ~ Q) w" Ul ~ 8 Q) 200 .;I ,.'.. .-< .-< K t"' 3 190 Q)" R " "0 180 -< ....0 ""Q) ' <11 1 1 70 " K~ """Q) " f.., <: " r-., \ s 0 ~ ~ ~ ~ ~ 1 60 "" 3-4 10-11 17-18 24-25 30- 31 Days of s tudy Figure 17 . Serum cholesterol and urinary neutral 17- ketosteroid and pregnanediol valu.es by days of s tudy for men • 8 §[pregnanediol []cholesterol ~neutral 17-ket osteroid 2 30 s ""rt ?, 7 2 20 ~ )< ::r 0" ) >--' 2 10 ~" >1 "0 >--' II 200 'g 4 ll ,.. >--' II 0 1/ 0 1/ 190 ~ It k B 1/ II 2 II l.80 II v l v v 170 v v v - )c ~ v ~ ~ ~ 0 II ~ ~ ~ -~ K v ~ 160 1- 3 4-6 11 l3 15 17 19-20 23- 25 26 - 28 Days of mens t rual cycle Table 18. Mean serum cholesterol and urinary neutral 17-ketosteroi d and p regna n~ d iol values by days of menstrual cycle for women 75 Table 16. The relationship between serum cholesterol and urinary neutral 17-ketosteroid, and pregnanediol by sex Sex Days of s tudy Neutral 17- Pregnanediol Cholesterol ketosteroid rng/24 hr urine mg/24 hr urine mg/100 rnl serum M 3-4 12.84 0.918 192 10-11 8.13 0 . 837 215 17-18 9.41 o. 786 227 24- 25 8 . 36 0 . 519 226 30-31 9.89 0 . 810 207 F Days of mens trual cycle 1-3 6.17 0.92 192 4-6 7. 26 0 . 91 194 9 5.37 0. 72 223 11 5.26 0.75 209 13 6 . 32 0. 78 213 15 7 .00 1. 25 207 17 7.37 1.18 217 19-20 4. 25 2.98 198 23-25 6 . 33 1. 93 214 26-28 5.75 l. 53 177 ~-I COMMENTS AND CONCLUSIONS The results herein reported are based upon the following assump tions: 1. The subjects kept a complete record of their food eaten, and ate the number of meals exactly as they should according to the experimental design . 2. The specimens were collected as outlined in the experimental design . The findings in this study suggested that the excretion of preg nanediol was affected by ovulation and the luteal phase during the period of the menstrual cycle. The excretion values started to increase after ovulation and had a peak in the luteal phase around the 22nd day of the menstrual cycle when the cycle was normal, that is , 28-30 days. As was expected, these data indicated that no sex difference in serum cholesterol level was found among these young normal adults with a mean age 21. The serum cholesterol values varied from day to day due to the dietary intake, physical activity , enviromental factors , or other factors. It was interesting that this study showed a reciprocal relationship between serum cholesterol and the sex hormones, neutral 17- ketosteroid in males and pregnanediol in females . From the results of this study, no definite relationship was observed between the number of meals as 2 versus 3 and serum total cholesterol level and pregnanediol excretion . If the small number of n subjects are considered as in this study, the possibility can not be ruled out that vith larger numbers of subjects and controlled diets, greater differences might have been observed and the effect of indivi dual difference vould not have been so pronounced. Reports in the literature have s uggested that 1ncreasing the number of meals might be beneficial to people in lowering the serum cholesterol levels, especially when trying to lose veight. It might be that increasing the number of meals to 5 or 6 times per day vould be helpful. Hovever, results in this study might be interpreted to indicate that total calories, fat content, or some other dietary constituent may be more important than the number of meals per day. Variation betveen individual subjects of the same sex vas great. Further work using a larger population is needed to clearly define the effect and relationships that might exist between number of meals and serum cholesterol level . Since the number of s ubj ects in this study was only 9, one high or low value could skew the data. Additional determinations of the serum cholesterol level s , and degradation products of sex hormones on a larger population of normal adults under regular living condition is recommended , to better establish the relationship of serum cholesterol, endocrine condition, and dietary factors. If possible, periodic observations of these same population in later adulthood is s uggested also, to define the above relationship to the incidence of atherosclerosis and coronary heart disease. SUMMARY Four men and five women, university students, served as s ubJects in a study of interrelationship among ur1nary pregnanediol, serum cholesterol, sex, and dietary factors while consum ing self-selected diets under ordinary living conditions. Two test periods of 31 days each were used. Half of the subjects ate two meals a day with no lunch and the other group had three meals a day in the first period. Both groups reversed their food pattern for the second period . The subjects maintained constant weight on their self-chosen d1ets during the entire studying periods. Calories, protein, carbohydrates, and fats of their diets were calculated. From the food data obtained , almost the same number of calories were consumed when the subjects ate two meals a day as when three meals a day were eaten. Finger-tip blood and urine specimens we r e collected tw1ce a week for the determination of cholesterol and pregnanediol, respectively. Additional urine samples were also collected on specified days during menstrual periods for female subjects. No sex difference, in free and total cholesterol values was noted. The mean values of free and total cholesterol were 54 . 5, 53 . 6, and 214 , 204 mg for men and women, respectively. The variation in weekly choles terol values shown by individual subjects was large in most cases; these differences were significant. For the men the differences between the low and high total cholesterol values in any one period ranged from 29 to 89 mg. The range for the women was even greater, 9 to 84 mg, but one 79 woman showed the least variation of any subject (13 and 25 mg for periods 1 and 2). Another woman had only 9 mg between low and high values during period 2 but had a difference of 51 mg in period 1. There was no definite evidence which showed that 2 or 3 meals per day had an effect on serum cholesterol levels or on pregnanediol values although mean pregnanediol values of all subjects were slightly higher when three meals were consumed than on two meals. Pregnanediol values were much higher for the women than for the men. However, the women only showed higher pregnanediol values during the second half of the menstrual cycle. The overall means for urinary preg nanediol output were found to be 0.80 and 1.43 (0.81 for the first half of the menstrual cycle, and 1 . 77 mg daily for the latter half of the cycle) mg daily for men and women. The range of pregnanediolexcretion values were found to be between 0.15 and 2.46 mg per day for men, and between 0 . 13 and 5.76 mg per day for women. The mean values of preg nanediol excretion for female during the latter half of the menstrual cycle were twice those of the male subjects . The mean values of all the subjec ts were slight higher when three meals were consumed than on two meals. The menstrual cycle did effect t he excretion of pregnanediol which rose in the second half of the men s trual cycle and precipitously fell prior to the onset of menstruation. Variation in serum cholesterol values were also influenced by ovulation. The concentration of cholesterol fell at the same time that the pregnanediol values showed an increase . Further work usi ng a greater number of subjects is recommended t o clearly define the relationship of serum cholesterol , endocrine cond i tion and dietary factors. LITERATURE CITED Adlersberg, D., L. E. Schaefer, A. G. Steinsberg, and C. I . Wang. 1956. Age, sex, serum, lipids and coronary atherosclerosis. Journal of American Medical Association 162(7):619-622. Adlers berg, D. 1958. Influences of hormones on circulating lipids. Hormone s and Atherosclerosis p. 301-313. Aftergood, L., and R. B. Alfin-Slater. 1965. Dietary and gonadal hormone effects on lipid metabolism in the rat. Journal of Lipid Research 6:287-294. Albrink, M. J. 1965. Diet and cardiovascular disease. Journal of the American Dietetic Association 46:26-29. Allen, W. M., and E. Viergiver. 1941. A titrimetric method for the de termination of sodium pregnanediol glucuronidate in the urine of pregnant women. Journal of Biological Chemistry 141:837-852. Allen, W. M. 1950 . Simple method for analyzing complicated absorption curves, of use in the colorimetric determination of urinary steroids. Journal of Clinical Endocrinology and Metabolism 10:71-83. Anonymous. 1961. Report by the American Heart Associations Central Committee for Medical and Community Program. What's New i n Home Economics 25:93-95. Astwood, E. B., and G. E. Seegar Jones. 1941. A simple method for the quantitative determination of pregnanediol in human urine. Journal of Biological Chemistry 141:837. Bang, H. 0. 1964. A simplified method for the quantitative determinat •on of pregnanediol in urine. Journal of Chromatography 14(3):520-523. Bloch, K. 1945. The biological conversion of cholesterol to pregnane diol. Journal of Biochemistry 157:661-666. Bohle, E., and R. Biegler. 1958. The relation between increased llpide, constitution, and age in arteriosclerosis. Medizinsche 664:70 . Bowes, A. deP., and C. F. Church. 1963. Food Values of Portions Commonly Used. J. B. Lippincott Co. , Philadelphia and Montreal. Boyd , E. M. 1937. The interaction of blood lipids. Canadian Journal of Research 15(D):l-25. 81 Cantarow, A. , and B. Schepartz . 1962. Biochemistry. Third edition. W. B. Sunders Comp any , Philadelphia, Pennsylvania. Cohn, C., D. Jos eph, and M. D. Allweiss . 1962. Nutritional effects of feeding frequency . American Journal of Clinical Nutrition 11 : 356-361. Cohn, C. 1961. Meal eating, nibbling and body metabolism. Journal of the American Dietetic Association 38:433. Cohn, C., R. Pick, and L. N. Katz. 1961. Effect of meal eating com pared to nibbling upon atherosclerosis in chickens. Circulation Research 9:139 . Cohn, C. 1964 . Feeding patterns and some aspects of cholesterol metabolism. Federation Proceedings 23:76-81. Co ok, R. F. 1958. Cholesterol. Acadimic Press Inc., Publishers, New York. p. 193, 194, and 202. Eberlein, W. R., and A. M. Bongiovanni. 1958. A paper chromatographic me thod for the measurement of pregnanediol in urine. Journal of Clinical Endocrinology and Metabolism 18:300-309. End e , N. 1962. Starvation studies, with special reference to cholesterol . American Journal of Clinical Nutrition 11 : 270-280. Fruton, J. S., and S. Simmonds. 1958. Biochemistry. John Wiley & Sons , Inc ., New York. p. 637-644. Galloway, L. S., P. W. Nielson, E. B. Wilcox, and E. M. Lantz. 1957. Microdetermination of cholesterol by use of 0.04 ml of blood serum. Clinical Chemistry 3(4):226- 232. Gertler, M. M., S.M. Garn, and E. F . Bland. 1950. Age, serum choles terol and coronary artery disease. Circulation 2:517. Gertler, M. M. , and S . M. Garn. 1950. Lipid interrelationshi p in health and in coronary artery disease. Science 112(1):14-16. Goldsmith, G. A. 1964. Therapy of hypercholesterolemia. The American Journal of Digestive Disease 9:651. Grollman, A. 1941. Essential of Endocrinology. Second edition. J. B. Lippincott Co., Philadelphia, London. p. 514-531. Hawk, B. B. , B. L. Oser, and W. H. Summerson . 1954 . Practical Physic logical Chemistry. Thirteenth edition. McGraw-Hill Book Co., Inc . , New York. p. 900. Hollifield, G. , and W. Parson. 1962. Metabolic adaptations to a "stuff and starve" f eeding program . I. Studies of adipose tissue and liver glycogen in rats limited to a short daily feeding period. Journal of Clinical Investigation 41:245-249. Jones , H. B. , J. W. Gofman, F. T. Lindgren, T. P. Lyon, D. M. Graham, B. Strisower , and A. V. Nichols. 1951. Lipoproteins in a theros clerosis. American Journal of Medicine 11:358-380. Keys, A. , and F . Fidanza. 1960 . Serum cholesterol and weight of cor onary patients. Circulation 22: 1091-1106. Keys , A., F. Fidanza, V. Scardi, and C. Be rgami. 1952 . The trend of serum cholesterol levels with age. Lancet 2:209-210. Keys, A., 0. Mickelsen, E. Miller, E. R. Ha yes , and R. L. Todd. 1950. The concentration of cholesterol in the blood serum of normal man and its relation to age. Journal of Clinical Investigation 29 : 1347-1353. Klepper, A., E. A. Michie , and J. B. Brown. 1955. A method for the determination of urinary pregnanediol. Journal Endocrinology 12:209-219. Kornerup, V. 1950 . Concentration of cholesterol, total fat, and phospholipid in serum. Archives of Internal Medicine 85:398-415. Kritchevsky, D. 1958. Cholesterol. John Wiley & Sons Co ., Inc. , New York, and London. p. 203. Krupp, S. 1960. Physician ' s Handbook. Eleventh edition. Lange. (Original not seen; Ab s tracted in Harper, H. A. 1961. Review of Physiological Chemistry, eighth edition. Lange Medical Publica tions, Los Altos, California . p. 132). Lawry, E. Y. , G. V. Mann, A. Peterson, A. P . Wysocki, R. O'Connell, and F. J . Stare. 1957. Cholesterol and beta lipoproteins in the serum of Americans. American Journal of Medicine 22:605-623 . Lewis, B. 1958 . Effect of certain dietary oils on bile acid secretion and serum cholesterol. Lancet 1:1090-1092. Mangold, H K. , and H. H. 0. Schmid . 1964. Thin-layer chromatography . Methods of Biochemical Analysis 12 :393-449. Okey, R. , and R. E. Boyden. 1927. Studies of the metabolism of women . III. Variations in the lipid content of blood in relation t o the menstrual cycle . Journal of Biological Chemistry 72 :261-281 . Okey, R., and D. Stewart. 1933. Diet and blood cholesterol in norma l women. Journal of Biological Chemistry 99:717-727. 83 Okey, R. , G. She1er, and R. Re1d. 1960 . Food restriction and cholesterol metabo lism. J ournal of American Dietetic Association 36:441-446 Oliver, M. F . , and G. S. Boyd. 1958. Hormonal aspects of coronary artery disease . Vitamins & Hormones 16:148-178. Randera h, Ku rt . 1963. Thin-layer chromatography. New York, Academic Press. Sperry, W. M. and M. Webb . 1950a. The effect of increasing age on serum cholesterol concentration . Journal of Biological Chemistry 187: 107-110. Sperry, W. M. and M. Webb. l950b. A revision of the Schoenheimer Sperry method for cholesterol determination. Journal of Biological Chemistry 187:97-106. Stamler, J. , D. M. Berkson, Q. D. Young, H. A. Lindberg, Y. Hall, L. Mojonnier , S. L. Andelman. 1963. Diet and serum l 1pids in atherosclerotic coronary heart disease. The Medical Cl1n1cs of North American 47:3. Venning, E. H. 1937. Gravimetric method for t he determination of sodium pregnanediol glucuronidate. Journal of Biological Chemistry 119:473-480. Waldi, D., and F. Munter. 1960. Early pregnancy test. Medic na Exp. 3:45 (Original not seen, abstracted in Kurt Randerath Thin-layer Chromatography p. 123-125. 1963. Academic Press., New York, and London). Watt, B. K. and A. L. Merrill. 1963. Composition of foods--raw, processed, prepared . U. S . Department of Agriculture. Agricul tural Handbook. No. 8 (Revised Edition). West, E. S. , and w. R. Todd. 1962 . Textbook of Biochemistry , Macmillan Co., New York. p . 496. Wilcox, E. B. , B. Hawthorne, R. Okey. 1962. Cholesterol problems. California Agricultural Experiment Station. Bulletin 785. p. 4 . Wilkens, J. A. , H. DeWit, and B. Bronte-Stewart. 1963 . A proposed mechanism for the effect of fats on serum cholesterol. Nutrition Review 21:42-45. Yu, S, H. 1964. Neutral 17-ketosteroid and 17, 21-dihydroxy-20-keto steroid in urine of university students . Unpublished MS thesis . Utah State University Library , Logan. p . 27. Zarrow, M. X., J . M. Yochin, and J. L. McCarthy. 1964. Endocrinology . Academic Press, New York. APPENDIX Table 17. Urinary 5 ~ -pregnane-Jot , 200( -diol values for individual subj ects consumi ng two meals and three meals per day Number Day of study Subjects Sex of meals Period 3- 4 10-11 17-18 24-25 30- 31 Mean values mg/24 hour urine B.W. M 1 0.626 0.705 0.552 0.447 0.799 0.626 2 0.858 0.923 0.821 0.995 0 . 899 D.P. M 1.139 0.688 1.407 0.917 1.038 0.801 0.429 0.351 0.152 0.413 0.429 H.A. M 0.437 0.354 0 . 213 0.339 0.563 0.381 0.496 0.575 0 . 656 0.598 0.338 0.533 B.E. M 2 2 1.005 1.083 0.789 0.806 0.446 0 . 826 3 1 1.976 1.931 1.495 0. 770 2 .005 1.635 T.D . F 1 0.655 0.924 0.939 2.621 3.833 l. 793 2 0.949 1.015 0.321 4.178 3.730 2.038 J.A. F 1 1. 209 1.194 1.493 0. 291 0.943 1.026 2 1.050 0.680 4 .302 2. 788 0.186 1.801 H.J. F 3.838 2.989 0.996 2.364 2.546 2.342 1.024 0.585 0.500 1.113 B. N. F 0.624 0.552 0.469 2.867 2.262 1.355 1 2.728 0.487 0. 771 1.023 2.353 1.472 S.R. F 2 0.893 0.344 1.371 0.977 0. 761 0 . 869 1 0.274 2.997 0.879 1.571 4.796 2.103 00 ln Table 18. Urinary pregnanediol excretion values f or t he individual subj ects consuming two meals and three meals per day by day of menstrual cycle Sub- Meals Days of menstrual cycle ject 1-3 4-6 11 13 15 17 19-20 23-25 26-28 29-30 36-37 T.D. 0.66 1.00 1.39 0.96 0.89 0.94 2.61 3.83 1.06 0.84 0.41 1.02 0.47 0.20 0.32 4.18 3.73 J.A. 1. 21 0.87 1.19 0.60 1.07 1.49 0.29 0.95 0.89 o. 74 0.49 0.53 0.83 4.30 2.79 M. J. 1.85 0.91 0.42 1.00 2.24 2.38 2.36 1.92 0.91 1.37 1.02 0.60 2.15 0.59 0.50 B.N. 0.63 0.53 0.57 0.20 0.62 0.47 2.87 2.27 2.23 0.49 1.08 0. 77 0.95 0.64 1.02 2.36 s. R. 2 0.90 0.34 0.23 0.14 0.50 1.37 0.98 0. 77 3 0.88 0.58 0.54 1.57 5.19 5.45 4.80 Table 19. Urinary pregnanediol excr etion values for the individual subjects by day of menstrual cycle Days of mens trual cycle Subject 1-3 4-6 11 13 15 17 19-20 23- 25 26-28 29-30 36- 37 T.D. 1.06 0 . 84 0.54 1.01 0.93 0.58 0 .61 2.56 3.17 3 .83 J .A. 1.03 0.88 0.97 0.55 0.80 1.16 2.30 2 . 79 M.J. 1.85 0.91 0.90 1.01 1.42 2.27 1.77 0.65 B. V. 0.63 0 . 49 0.81 0 . 67 0.58 0.63 0.75 2 . 53 2 . 27 2.23 S. R. 0 . 90 0.88 0.46 0.39 0.86 2.85 3.41 4.80 0.98 0. 77 Table 20 , Serum t ot al chol e ste rol va l ues f or i ndividual subjects eating two me als and t hr ee meals per day Sub Sex No. of Period jects meals 10 11 24 25 30 31 Mean mg / 100 ml s erum B.W. M 193 186 246 213 213 250 26 7 243 22 6 210 212 242 233 214 227 22 3 D.P. M 195 201 230 227 217 232 216 240 236 22 2 214 194 237 201 256 240 280 202 179 223 H.A. M 2 2 185 206 177 183 200 193 198 201 137 187 3 1 171 172 175 223 223 225 224 213 227 206 B. E. M 199 189 236 205 194 152 206 170 160 190 171 182 209 204 199 230 228 180 235 204 T.D. F 1 189 198 177 225 215 199 198 196 197 199 2 211 189 202 175 195 199 214 184 210 198 J.A. F 2 1 181 197 251 244 235 223 261 231 196 224 3 2 254 242 218 252 217 195 227 252 232 M.J. F 2 1 185 199 229 213 278 251 247 195 201 222 3 2 212 172 274 197 219 210 169 211 208 B.N . F 2 171 190 217 190 261 201 120 150 210 197 3 1 156 130 144 222 221 231 199 227 157 187 S .R. F 17 3 189 185 181 190 187 180 165 194 183 1 154 156 192 220 151 194 178 175 156 175 co co Table 21. Serum free cholesterol values for individual subject s consuming two meals and three meals per day Sub- Sex No. of Period ject meals 10 11 24 25 30 31 Mean mg/100 ml serum B.W. M 47.3 46.9 75.9 46.5 62.5 61.0 59.9 63.1 57.9 60.5 51.4 57.8 61.0 53.5 64.6 56.1 D. P . M 2 1 45.3 51. 9 54 . 0 53.7 50.0 55.1 47.8 59.1 56.9 52.6 3 2 57.8 56.4 55.7 57 . 7 63.8 52.3 83.2 48 .0 49.6 58.3 H. A. M 2 2 50.9 61. 3 51.4 49.0 43.5 45.2 48.1 49.1 35.8 48.3 3 1 47.7 48 . 0 51.1 51.9 56.7 58.5 66 . 6 45 . 9 58.1 53.8 B. E. M 55 . 1 52.1 54 . 9 52 . 3 48.2 34.8 50 .0 48 . 1 50.6 49.6 1 45 . 5 49 .0 52. 2 43.1 45.4 49.1 57 . 6 48.9 55.6 49.6 T.D. F 1 42.0 48. 1 47 . 4 47.4 64 . 0 49.1 49 . 0 50.6 51.1 49.9 2 55.5 48 .6 61.1 49.4 50.8 47.3 60.5 47.9 50.9 52.4 J .A. F 2 1 38.7 49 .0 80. 1 72. 9 65 . 8 45.9 68.9 62 . 5 59.6 60.4 3 2 76.9 73.6 60.3 59.4 59.9 55.7 53.9 72.4 64.0 M. J . F 1 45 . 1 52 .4 54.5 46.4 69.6 56.3 68.3 53.1 50.0 55.1 2 60 . 3 40.8 67.4 56.4 53 . 0 60.1 47 . 9 64.0 56.2 B.N . F 2 43.9 51.1 53 . 9 44 . 8 77.1 50.5 46.4 42.3 48.9 51.0 1 39.9 40 . 7 36.8 57.4 54.0 61.1 57.3 50.9 38.4 48.5 S. R. F 47.3 50.8 59.3 44 . 9 44.6 41.2 41.9 45.7 46.0 46.9 1 46.6 50.6 53 .0 52 0 7 44.7 56.8 51.3 44.3 46.4 49 . 6 00 "'